Alka Patel - Iran To India - The Shansabānīs of Afghanistan, C. 1145-1190 CE-Edinburgh University Press (2021)

IRAN TO
INDIA
PATEL 9781474482226 PRINT.indd 1 14/09/21 5:36 PM

IRAN TO
INDIA
The Shansabanis of Afghanistan,
c. 1145–1190 ce
Alka Patel

Edinburgh University Press is one of the leading university presses in the UK. We publish
academic books and journals in our selected subject areas across the humanities and social
sciences, combining cutting-edge scholarship with high editorial and production values to
produce academic works of lasting importance. For more information visit our website:
edinburghuniversitypress.com
© Alka Patel, 2022
Cover image: Pigeon Tower, Injil, 7.5 km south of central Herat, Afghanistan © Didier
Tais 2011
Cover design: Andrew McColm
Edinburgh University Press Ltd

The Tun– Holyrood Road
12 (2f) Jackson’s Entry
Edinburgh EH8 8PJ
Typeset in 11/13 Adobe Garamond by

Servis Filmsetting Ltd, Stockport, Cheshire,
and printed and bound in Malta by Melita Press
A CIP record for this book is available from the British Library
ISBN 978 1 4744 8222 6 (hardback)

ISBN 978 1 4744 8224 0 (webready PDF)
ISBN 978 1 4744 8225 7 (epub)
The right of Alka Patel to be identified as author of this work has been asserted in
accordance with the Copyright, Designs and Patents Act 1988 and the Copyright and
Related Rights Regulations 2003 (SI No. 2498).
Published with the support of the University of Edinburgh Scholarly Publishing

Initiatives Fund.
Contents
List of Figures vi
Acknowledgments xii
Note on Dates and Transliteration xiv
Map of Imperial Ambitions xv
Map of Historical Ghur and Contiguous Regions xvi
Genealogy of the Shansabanis of Afghanistan xvii
Introduction: The Elephant and its Parts 1

1. Kingly Trajectories 40
2. Beginnings 79
3. The Early Shansabanis: Firuzkuh, Bamiyan, and Tiginabad/
Old Qandahar, c. 1140s–1170s 132
4. One and Several: Gharjistan, Chisht, and Imperial Firuzkuh 190
5. The “Ports of India”: Ghazna and Bust-Lashkari Bazar 237
6. Encountering the Many “Indias” 277
Epilogue: Iran to India 338
Appendix: Shansabānī Religious and Historical Inscriptions in

Afghanistan and Pakistan 342
Works Cited and Bibliography 357
Index 391

Figures
I.1 Remains of Shah-i Mashhad madrasa, Gharjistan 2

I.2 Minaret of Jam, general view from northeast 3
I.3 Tiginabad/Old Qandahar, grave with brick vault 4
I.4 Shahr-i Zuhhak (fifth–thirteenth centuries), general view 5
I.5 Sarkhushak, Bamiyan Province, Building A 6
I.6 Ghazni, palace of Mas‘ud III (1099–1115), excavation site,
courtyard 6
I.7 Bust, arch and city walls 7
I.8 Pictorial view of part of an early Sultanate palace
complex 7
I.9 Wall fragments and piers excavated at Lalkot 8
I.10 Mosque known as “Arhai Din-ka Jhonpra,” founded
c. 1199 ce 9
I.11 Mosque of Kaman, Rajasthan, India, founded c. 1195 10
I.12 Congregational Mosque, Herat, Afghanistan, c. 1200 ce,
exterior 11
I.13 Congregational Mosque, Herat, interior 12
I.14 Tomb of Ghiyath al-Din Muhammad ibn Sam, Herat, prior
to destruction in 1944 13
I.15 Tomb of Ghiyath al-Din Muhammad ibn Sam, Herat,
reconstructed c. 1940s 14
I.16 Tomb of Salar Khalil, a.k.a. ziyarat Baba Khatim, Juzjan
Province, Afghanistan 24
I.17 Tomb of Salar Khalil, interior 25
I.18 Tomb of Shahzada Sheikh Husain, Helmand Province,
Afghanistan, c. 1200 ce 26
I.19 Tomb of Shahzada Sheikh Husain, interior 27
I.20 “The fortress and citadel of Ghazni (Afghanistan),” James
Atkinson, watercolor, 1839 28
I.21 “Ruins of Old Kandahar Citadel,” albumen print, c. 1880–81,
Benjamin Simpson 28
I.22 Shahr-i Gholghola, general view, citadel, eleventh–thirteenth
century 29
I.23 Herat-Ghuriyan highway, Herat Province, Afghanistan 30

Figures vii
1.1 Congregational Mosque, Isfahan, Iran, tenth century ce;

exterior, east 49
1.2 Congregational Mosque, Isfahan, 1086–1087 ce; exterior,
north dome 49
1.3 Congregational Mosque, Isfahan, c. 1090 ce; interior, south
dome chamber 50
1.4 Minar known as Chihil Dukhtaran, c. 1100 ce, Jubara
quarter, Isfahan 50
1.5 Chihil Dukhtaran minar, historical inscription panel 51
1.6 Chihil Dukhtaran minar, topmost epigraphic bands 52
1.7 Mihrab of mosque at Sin, north of Isfahan, c. 1132 ce 53
1.8 Minar of mosque complex, Sin, c. 1134–1135 ce 53
1.9 Sin minar, topmost epigraphic bands 54
1.10 Robat Sharaf, near Sarakhs, Razavi Khurasan Province, Iran,
1114–1115, 1154–1155 ce 55
1.11 Robat Sharaf, second courtyard 56
1.12 Robat Sharaf, oratory 56
1.13 Robat Sharaf, intrados of arch with stucco decoration 57
1.14 Palace of Mas‘ud III, 1099–1115, Ghazni (excavation site) 58
1.15 Ghazni: minaret of Mas‘ud III, 1099–1115; and minaret of
Bahram Shah, 1118–1152 58
1.16 The Buddhist site of Tapa Sardar, near Ghazni, Afghanistan,
third–ninth century ce 59
1.17 Vaulted corridor (Ilkhanid) north of tomb of ‘Ala’ al-Din
Muhammad of Bamiyan, d. at Ghazna 1206 ce 66
2.1 Ahangaran, Ghur Province, Afghanistan 81
2.2 Jam-Firuzkuh, minar and confluence of Hari Rud and Jam
Rud 85
2.3 Ahangaran, surroundings and banks of Hari Rud 86
2.4 Detail of Herberg’s fieldwork map of central and southern
Ghur, c. 1970s 88
2.5 Yaman, Ghur Province, Afghanistan: ruins of a square tower
and cruciform “military structure” 89
2.6 Ghor Province, Yahan, fortification line and ruins 89
2.7 Ghor Province, Male Alau: brick tower, ruin, distant view 91
2.8 Ghor Province, Alana Valley: fortification line, ruins, distant
view 92
2.9 Taywara vicinity, Ghur Province: watch tower 93
2.10 Between Pasa Band and Dahane Nawrak: cruciform interior
of a defensive structure 94
2.11 Taiwara-Farah Rud 95
2.12 Ahangaran 95
2.13 Ghor Province, Yahan: fortification tower, ruin 96
2.14 Southern Ghur, southeast façade of a structure 97

viii Iran to India
2.15 Qala‘-yi Chahar Baradar, Ghur Province 98

2.16 Ghor Province, Male Alau: brick tower with stone base, ruin 99
2.17 The “lost” minar of Qala‘-yi Zarmurgh, c. 75 km south of
Jam-Firuzkuh 100
2.18 Minar of Qasimabad, Sistan 101
2.19 Balkh, minaret of Daulatabad (1108–1109); exterior, full view 102
2.20 Detail of Figure 2.16: Male Alau, Ghor Province, brick
tower, lower decoration 102
2.21 Damghan, Semnan Province, Iran: stucco dado with
palmettes 103
2.22 Nizamabad, Iran: wall fragment with stucco decoration 104
2.23 Detail of Figure 2.16: Male Alau, Ghor Province 104
2.24 Taq-i Bustan, central Iran: column capital 105
2.25 Minarets of Ghazni, distant view 110
2.26 Nad-i Ali, Zaranj, Sistan: remains of tower 111
2.27 Remains of baked-brick tower, Khwaja Siyah Push, Sistan 112
3.1 Jam-Firuzkuh, Ghur Province, Afghanistan: remains of
paved floor of probable mosque 138
3.2 Jam-Firuzkuh: remains of Qasr-i Zarafshan 140
3.3 Jam-Firuzkuh: remains of defensive towers 141
3.4 Jam-Firuzkuh: baked brick, plaster-lined cistern at Kuh-i
Khara 142
3.5 Jam-Firuzkuh: minaret of Jam, general view 144
3.6 Detail of Figure 3.4: mihrab relief on east face of
Jam-Firuzkuh minaret 145
3.7 Bamiyan: north cliff, eastern end with 35-m Buddha 147
3.8 The Bamiyan basin 149
3.9 Shahr-i Zohak, Bamiyan region, fifth–thirteenth century:
general view, triangular plateau with fortifications 151
3.10 Shahr-i Zohak: general view, crenellated tower and walls 152
3.11 Shahr-i Zohak: fortress ruins, interior 153
3.12 Shahr-i Gholghola, Bamiyan region, eleventh-thirteenth
century: general view, citadel 156
3.13 Shahr-i Gholghola: recovered wooden door leaves 157
3.14 Yakaulang, Saighan, and Kahmard Districts, western
Bamiyan Province, Afghanistan 158
3.15 Qala‘-yi Gauhargin, western Yakaulang District, Bamiyan
Province 158
3.16 Bamiyan vicinity, structural remains 159
3.17 Killigan, eastern Yakaulang District, Bamiyan Province 160
3.18 Chehel Burj, eastern Yakaulang District, Bamiyan Province 161
3.19 Barfak II, a.k.a. “Danestama,” c. 80 km northeast of
Bamiyan: plan 162
3.20 Barfak II/“Danestama:” east side of central courtyard 163

Figures ix
3.21 Barfak II/“Danestama:” mihrab from southeast 164

3.22 Barfak II/“Danestama:” detail of stucco revetment in
mihrab 165
3.23 Sarkhushak, north of Bamiyan 166
3.24 Sarkhushak, Building A: plans and elevations 166
3.25 Sarkhushak, Building E: plan and elevation 167
3.26 Sarkhushak, Building A: vault 168
3.27 Qandahar, Afghanistan: Qaitul stupa and citadel of Old
Qandahar 170
3.28 Old Qandahar cemetery, Burial Mound A 172
4.1 Remains of Shah-i Mashhad madrasa, c. 1165–1175 ce,
Gharjistan, Afghanistan 196
4.2 Shah-i Mashhad madrasa and Murghab River 200
4.3 Shah-i Mashhad madrasa on Google Earth with
superimposed plan 200
4.4 Shah-i Mashhad madrasa: south façade 202
4.5 Shah-i Mashhad madrasa: possible tomb at southeast
corner 202
4.6 Shah-i Mashhad madrasa, c. 1165–1175 ce: south façade 203
4.7 Shah-i Mashhad madrasa: southern monumental entrance 204
4.8 Shah-i Mashhad madrasa: south façade 205
4.9 Shah-i Mashhad madrasa: north entrance 206
4.10 Ghazni, minaret of Mas‘ud III (1099–1115), with citadel in
background 208
4.11 Ghazni, minaret of Mas‘ud III (1099–1115), middle section 209
4.12 Chisht, mausoleums or madrasa/mosque complex, Herat
Province, Afghanistan: Mausoleum A, exterior 211
4.13 Chisht: Mausoleum A, interior 212
4.14 Chisht: Mausoleum A, interior, inscription 213
4.15 Chisht: Mausoleum A, interior, arch 215
4.16 Chisht on Google Earth 216
4.17 Chisht: Mausoleum B, interior, mihrab 216
4.18 Jam-Firuzkuh, Ghur Province, Afghanistan: minaret of Jam,
general view 218
5.1 Ghazna, Ghazni Province, Afghanistan: plan from aerial
photograph of western part of city 243
5.2 Ghazna, central palace: plan of excavated areas and
hypothetical reconstructions 244
5.3 Ghazna, central palace: fragment of painted stucco
decoration, detail 1 245

x Iran to India
5.6 Ghazni, palace of Mas‘ud III (1099–1115): dado remnants 246

5.7 Detail of Figure 5.6: juxtaposition of Ghaznavid- and
Shansabani-period revetments 247
5.8 Bust, Helmand Province, Afghanistan; general plan 248
5.9 Bust, citadel and arch: general view from northeast 249
5.10 Bust, citadel: seven-story well, arches 250
5.11 Bust, citadel: seven-story well interior shaft 251
5.12 Bust, arch, south pillar, carved decoration 252
5.13 Bust, arch: front face and intrados, north 253
5.14 Lashkari Bazar, Helmand Province: South Palace, plan of
excavations 255
5.15 Lashkari Bazar, South Palace, southern façade 256
5.16 Lashkari Bazar, Summer (or Grand) Palace, exterior 256
5.17 Lashkari Bazar, South Palace: south portal detail 257
5.18 Lashkari Bazar, South Palace: Mosque of the Forecourt 258
5.19 Lashkari Bazar, South Palace: courtyard 258
5.20 Laskhari Bazar, South Palace: Oratory F, north of central
courtyard 259
5.21 Lashkari Bazar, South Palace: Oratory F, stucco decoration 260
5.22 Lashkari Bazar, South Palace: Oratory F, mihrab 261
5.23 Lashkari Bazar, Palais aux Raquettes, exterior 265
5.24 Lashkari Bazar, South Palace: Audience Hall I 266
5.25 Lashkari Bazar, South Palace: Audience Hall I, detail of
stucco revetments 267
5.26 Lashkari Bazar, South Palace: Audience Hall I, detail of
paintings 268
5.27 Lashkari Bazar, South palace, view looking south 269
5.28 Lashkari Bazar, general view 270
6.1 Buddhist monastic complex, Takht-i Bahi,
Khyber-Pakhtunkhwa Province, Pakistan 284
6.2 Rajagira Mosque, Udegram, Khyber-Pakhtunkhwa Province,
Pakistan 286
6.3 Rajagira Mosque: mihrab 287
6.4 Temple, Kalar, West Panjab, Pakistan; exterior façade 288
6.5 Temple, Mari Indus, West Panjab, Pakistan 289
6.6 Large temple, Ambh Sharif, West Panjab, Pakistan 290
6.7 Temple, Nandana, West Panjab, Pakistan; interior 291
6.8 Large temple, Ambh Sharif, exterior: detail of shikhara 291
6.9 Mosque remains, Dibul-Banbhore, Sindh, Pakistan 292
6.10 Tomb attributed to Muhammad ibn Harun, Lasbela,
Baluchistan, Province Pakistan 293
6.11 Congregational Mosque and Thamban Wari Masjid, Jam
Jaskars Goth, Sindh Province, Pakistan; plans, elevations,
drawings 294

Figures xi
6.12 Chhoti masjid, Bhadreshvar, Kachch District, Gujarat, India;

interior 295
6.13 Tomb known as Shrine of Ibrahim, Bhadreshvar; interior 296
6.14 Temple, Bhodesar, Nagarparkar District, Sindh Province;
exterior base of shikhara 297
6.15 Multan, West Panjab, Pakistan; tomb of Sheikh Yusuf
Gardizi 301
6.16 Riba† of Ali ibn Karmakh, Kabirwala, near Multan, West
Panjab, Pakistan; plan 304
6.17 Riba† of Ali ibn Karmakh 305
6.18 Riba† of Ali ibn Karmakh: interior, mihrab 306
6.19 Riba† of Ali ibn Karmakh: framing bands of mihrab, in situ
c. 1980s 307
6.20 Riba† of Ali ibn Karmakh: detail of epigraphic bands framing
mihrab 308
6.21 Riba† of Ali ibn Karmakh: detail of reused fragment 309
6.22 Riba† of Ali ibn Karmakh: detail of pilaster framing mihrab 310
6.23 Riba† of Ali ibn Karmakh: detail of base moldings of mihrab
hood 311
6.24 Tomb of Sadan Shahid, Muzaffargarh, near Multan, West
Panjab, Pakistan; exterior, west façade 311
6.25 Tomb of Sadan Shahid; east façade 312
6.26 Tomb of Sheikh Sadan Shahid, interior 312
6.27 Tomb of Sadan Shahid: exterior, detail of decorative niche 313
6.28 Tomb of Sadan Shahid, north half of east façade 313
6.29 Tomb of Ahmad Kabir, Dunyapur, near Multan, West
Panjab, Pakistan; east façade 314
6.30 Tomb of Ahmad Kabir, from northwest 314
6.31 Tomb of Ahmad Kabir, interior 315
6.32 Tomb of Ahmad Kabir: exterior, detail of decorative niche 316
6.33 Tomb of Duagan, Sukkur, Sindh Province, Pakistan 317
6.34 Tomb of Duagan; exterior, detail of pilasters 318
6.35 Chhoti Masjid, Bhadreshvar Kachh District, Gujarat, India;
detail of pilaster capital 318
6.36 Shrine of Ibrahim, Bhadreshvar; foreporch 319
6.37 Lal Mara Sharif, near Dera Ismail Khan, West Panjab,
Pakistan; tombs at necropolis 320
6.38 Lal Mara Sharif, Tomb IV, west façade 322
6.39 Lal Mara Sharif, Tomb III, west façade 325
6.40 Lal Mara Sharif, Tomb I, west façade 326
6.41 Lal Mara Sharif, Tomb IV, west façade 326

Acknowledgments
The completion of any project is possible only because of the belief, sup-
portiveness, and generosity of many. And an uncommon one such as
this – involving more than one book – will, I am afraid, require trespassing
on the aforementioned qualities long after the acknowledgments for this
book are penned and in print. It is no exaggeration to say that the whole
endeavor would not have reached this sar manzil without the unstinting
contributions of others. I humbly and gratefully acknowledge the entities
that have eased the way thus far, and the individuals who have embarked
on this journey with me.
The project received its first boost with an American Institute for
Indian Studies Senior Fellowship for completion of in-depth architec-
tural documentation in north India (first begun during doctoral fieldwork
throughout India and Pakistan). Thereafter, fellowships from the American
Philosophical Society and the UC Office of the President supported
fieldwork in Iran and Afghanistan. Iranian warmth and hospitality are
world-renowned for good reason; I am particularly grateful to Professor
Javad Abbasi (Firdausi University) for sharing his time, knowledge, and
conversation. In Afghanistan, neither work nor the trip itself would have
been possible without the guidance and support of Praveen Kumar KG of
the Indian Embassy in Kabul; Dr. Daud Shah Saba, then Governor of Herat
Province; Mr. Tarachand, Consul General of India in Herat; and Phillipe
Marquis of the Délégation Archéologique Française en Afghanistan.
After foraging for and gathering vast amounts of data, its collation
and composition into a narrative necessarily calls for cloistering, pref-
erably with responsive interlocutors who help one feel less alone (and
less crazy). Fellowships at the Getty Research Institute (GRI; 2017–18)
and at the National Humanities Center (NHC; 2019) both provided the
perfect balance of solitude for writing and meaningful conversations with
wonderfully smart people. I cannot thank the GRI and the NHC enough;
these opportunities renewed my own faith in the project and were instru-
mental in completing the book manuscript of this first “installment.”
Most recently (April 2020), the Dean’s Office of the School of
Humanities at the University of California, Irvine (UCI) provided
the unique opportunity of a workshop on the manuscript prior to its

Acknowledgments xiii
finalization and submission to a press for review. It is worth stating in no

uncertain terms that, at this stage in the work, the feedback from respected
senior colleagues in such diverse fields as Anthropology, Archeology, Art
History, and History – and spanning virtually all of the second mil-
lennium ce across western, southern, and central Asia – made the final
product that much stronger. Surely, such a practice would be welcomed
by all authors, not only for the assured improvement of their books, but
also for the enduring conversations that are generated in such well-defined
but informal contexts.
Supportive entities are, of course, made up of individuals. The abso-
lutely top-notch librarians and research assistants at my home campus
of UCI, the GRI and the NHC have earned my enduring thanks.
Additionally, I extend my heartfelt gratitude to the students in the 2018
Getty Consortium graduate seminar – you all remained steady through-
out the whirlwind engagement with Iconoclasm – and to the colleagues
who brought fresh perspectives to our sessions. Among the latter were
fellow Getty Scholars who patiently witnessed a novice’s forays into topics
as varied as colonial-era photography and the anthropology of nomad-
ism, and to whom I am deeply grateful for their gentle guidance: John
Falconer, Tim Murray, and Norman Yoffee.
At UCI and its vicinity, several colleagues – many of them also friends –
deserve thanks as stalwart allies in this continuing project: Maureen Burns,
Matthew Canepa, Touraj Daryaee, Saied Jalalipour, Sara Mashayekh,
Laura Mitchell, Kavita Philip, Khodadad Rezakhani, and Amanda Swain.
I am especially grateful to Amanda and Saied for their flawless execution
of the aforementioned manuscript workshop, which brought together a
roster of scholars whose generosity in engaging with the work and the
insights they offered were truly humbling: Mitchell Allen, Catherine
Asher, Warwick Ball, Thomas Barfield, Will Chamberlain, Richard Eaton,
Vivek Gupta, Beatrice Manz, Fatima Quraishi, and Layah Ziaii-Bigdeli.
Adding to this already lengthy list of all who have a stake in this project,
I happily include the following colleagues and friends for their unfailing
help and support: Viola Allegranzi, Warwick Ball (yes, twice!), Roberta
Giunta, Michael O’Neal, Jawan Shir Rasikh, Franziska Seraphim, Sunil
Sharma, and David Thomas. By reading and commenting on chapter
drafts, providing otherwise unobtainable research materials, and overall
staying the course through my own fluctuating assessments of the value
of my work (thank you again, David!), you were instrumental in making
this happen.
Last, yet never least, my family: Didier, Anika, Max (you are missed),
Christine S., and the Tarboxes – by your very presence, you make me
better as a person, friend, mother, partner, scholar.
With such good fortune in colleagues, friends and family, I emphasize
that any shortcomings in this work are my own.

Note on Dates and Transliteration
For accessibility and ease of reading the main text of this book, every
effort has been made to make dates and transliterations less cumber-
some. Rulers’ reigns and timespans of specific years are provided in both
Hijra and Common Era dates; while less specific ranges are given only in
the Common Era. Furthermore, transliteration has been utilized where
strictly called for: the many Arabic, Persian and Indic-language words
throughout the main text and footnotes have not been transliterated; only
the ayn and hamza have been differentiated. However, to be consistent
with the extensive Arabic and Persian texts of inscriptions reproduced in
the Appendix, all non-English words have been transliterated there: the
Arabic–Persian adhering to the International Journal of Middle East Studies
recommendations, and the Indic-language words following the guidelines
in John T. Platts’s A Dictionary of Urdu, Classical Hindi and English.

Map of Imperial Ambitions. © Didier Tais 2020.

Anbir/Sar-i Pul
Map of Historical Ghur and Contiguous Regions. © After Michael O’Neal 2015.

PATEL 9781474482226 PRINT.indd 17
Genealogy of the Shansabanis of Afghanistan. © By kind permission of Michael O’Neal 2013.
14/09/21 5:36 PM
Introduction:
The Elephant and its Parts
There was a great city in the country of Ghur, in which all the
people were blind. A certain king passed by that place, bringing
his army and pitching his camp on the plain. He had a large and
magnificent elephant to minister to his pomp and excite awe,
and to attack in battle. A desire arose among the people to see
this monstrous elephant, and a number of the blind, like fools,
visited it, every one running in his haste to find out its shape and
form. They came, and being without the sight of their eyes groped
about it with their hands; each of them by touching one member
obtained a notion of some one part; each one got a conception of
an impossible object, and fully believed his fancy true.1
T his version of the well-known parable of the blind men and the
elephant comes from the long mathnavi titled Hadiqat al-Haqiqa,
the ambitious last work of the poet Majdud ibn Adam Ghaznavi
(c. 1087–1130/1140 ce), best known by his ‘urf, Sana’i.2 It is a fitting point
of departure for the ensuing project on the Shansabanis of Ghur, or the
“Ghurids” as they are commonly denoted in modern scholarship (but not
here, for reasons discussed below). Sana’i not only located the didactic
story in Ghur – whence unfolded the historical processes plotted in these
two volumes – the parable itself appears to have been transposed from
the Buddhist Udana, among the earliest compilations of the Enlightened
One’s solemn utterances.3
In his intellectual inspirations for Hadiqat, Sana’i enacted the per-
ennial continuity of sources and realities,4 many defined as non-Islamic
and thus frequently understudied by modern “Islamicists” and others as
they investigate cultural productions emerging from the vast geography
of Islamic-Persianate Eurasia. Surely the poet would have been pleased to
know that, almost a millennium after his own lifetime, his selection5 of
the elephant parable still provides a compelling homily: for our purposes,
Sana’i succinctly and humorously exposed our figurative blindness, born
of our temporal distance from the historical pasts we are trying to access –
a blindness that is arguably compounded by the increasingly narrow schol-
arly specializations mobilized to attempt such recoveries.

2 Iran to India
These volumes are about encountering a proverbial elephant – here

symbolizing the seemingly sudden emergence in the mid-twelfth century
of an obscure group from central Afghanistan into historical retrievability
as the “Ghurids” – more accurately the Shansabanis (cf. below); and their
even more amazing rise as they fashioned an empire extending from Herat
in modern western Afghanistan to the Brahmaputra River in northeastern
India–western Bangladesh. To date, the sequence of events leading to
Shansabani ascendancy and the impetuses behind them are only partially
known, however, the contemporaneous textual sources offering sometimes
contradictory information or eliding them altogether (cf. Chapter 1).
Furthermore, Shansabani architectural traces have been analyzed piece-
meal, thus presenting a seemingly incomprehensible pastiche, not unlike
Sana’i’s elephant when perceived only through its individual parts. The
present study essentially undertakes an architectural biography (see below):
the two volumes bring together the corpus of Shansabani-patronized
structures as among the primary historical sources with much to contrib-
ute toward our understanding of the Shansabanis, who were not only one
of the “nomadic dynasties” emerging in Eurasia6 as of the early second
millennium ce, but an imperial formation impacting the subsequent tra-
jectories of Central and South Asia.
Patronage of public or “civic” architecture such as madrasa complexes
(Figure I.1) has been attributed to the Shansabanis, but their “capital” at
Figure I.1 Remains of Shah-i Mashhad madrasa, Gharjistan (Badghis Province,

Afghanistan). © Photograph Bernt Glatzer c. 1970.

Introduction 3
Figure I.2 Jam, minaret of Jam (c. 1180) [sic], general view from northeast, with steep
valley in background. © Photograph Josephine Powell c. 1960. Harvard Fine Arts Library,
Special Collections.
Jam-Firuzkuh in central Afghanistan (Figure I.2) long remained enigmatic

due to its isolated location and lack of palatial remains; even more so
was the Shansabanis’ exclusively funerary presence in the strategically and
commercially important region of Qandahar (Figure I.3). Their mate-
rial footprint in the Bamiyan valleys (Figures I.4 and I.5; Figure I.22) –
unexamined as a whole, until now (see Chapter 3) – is admittedly difficult
to discern, given the mixture of reused and newly constructed traces,
both on a significant scale. Even the momentous Shansabani conquest of
Ghazna and Bust-Lashkari Bazar in c. ah 568–569/1173 ce resulted largely
in renovations of existing palatial areas (Figure I.6), with only punctually

4 Iran to India
Figure I.3 Tiginabad/Old Qandahar, grave with brick vault. From Whitehouse
1976: fig. IIa.
built new monuments (Figure I.7). Notably, it appears to have been their
entry into the north Indian duab in the later ah 580s/1190s ce that sig-
naled new, monumental patronage in palatial (Figures I.8 and I.9) as well
as religious ambits (Figures I.10 and I.11) – both replete with identifiably
reused building materials. At the same time, in Herat the Shansabanis left
distinct traces within the renowned, centuries-old congregational mosque
complex (Figures I.12 and I.13), addorsing to its northeastern perimeter the
tomb (no longer extant) of Sultan Ghiyath al-Din Muhammad ibn Sam
(r. ah 558–599/1163–1203 ce; see Volume II) (Figures I.14 and I.15).
Although the imperial Shansabanis achieved and maintained their
greatest territorial extent for only about two decades (c.1192–1210 ce),
this was nonetheless a feat rarely seen for almost a millennium: in the first
centuries of the Common Era, the Kushanas brought ancient Bactria,

Introduction 5
Figure I.4 Shahr-i Zuhhak (fifth–thirteenth centuries), general view, view along
crenellated fortifications, with valley in background. © Photograph Josephine Powell
c. 1960. Harvard Fine Arts Library, Special Collections.
Gandhara, the Panjab, and the Ganga-Yamuna duab under their aegis.
Thereafter, the Yamini-Ghaznavids (c. ah 365–582/975–1186 ce) con-
ducted plundering raids into the duab but with minimal traces of an
imperial presence there.7 The Kushanas’ imperial expansion had had
important consequences for all of Eurasia throughout the first millennium
ce.8 Nearly a thousand years later, the Shansabanis created a similar con-
federate empire (see esp. Chapters 1, 3–5, this volume), extending from the
western Hari Rud through the Kuh-i Baba range in modern Afghanistan,
to the Sulaiman Mountains and the Indus valleys of modern Pakistan,
and ultimately reaching the northern plains of modern India. In the foot-
steps of the Kushanas (rather than the Ghaznavids), then, the Shansabanis
re-conjoined the Indic and Iranian cultural–geographical worlds and
unleashed far-reaching consequences, this time throughout the second
millennium ce into the present day.
As further examined in these volumes, two among these consequences

6 Iran to India
Figure I.5 Sarkhushak, Bamiyan Province, Building A. From Baker and Allchin
1991: fig. 5.15.
Figure I.6 Ghazni, palace of Mas‘ud III (1099–1115), excavation site, courtyard.
Marble floor and dado remains. © Photograph Josephine Powell c. 1960.
Harvard Fine Arts Library, Special Collections.

Introduction 7
Figure I.7 Bust, arch and city walls. © Photograph Josephine Powell c. 1960.
Figure I.8 Pictorial view of part of an early Sultanate palace complex, exposed
during the ASI excavations conducted 1992–1995. From Balasubramaniam 2005: 110,
fig. 30.

8 Iran to India
Figure I.9 Wall fragments and piers excavated at Lalkot. © Courtesy of the
Archaeological Survey of India, New Delhi.
are particularly noteworthy. First, during the 1190s ce the Shansabanis’

successful campaigns into the north Indian duab saw the establishment
in the region of their interpretation of Perso-Islamic cultures of power9
(cf. Chapter 1), to be contrasted to the Muslim mercantile presences
along the subcontinent’s coasts and some inland nodes dating from the
eighth century onward.10 Second, the Shansabani campaigns instigated
a broad geo-historical momentum, which lasted through the early six-
teenth century: successive waves of Islam-affiliated warlords entered the
duab along the self-same northwestern corridors and ruled the region
with varying longevities. This span of a bit more than three centuries –
essentially from the 1190s until the arrival of the Timurid prince Babur
(ah 888–936/1483–1530 ce) in 1525 – is generally bracketed in modern
scholarship as the Sultanate period.11
But, despite four Persian-language monographs within the last fifty
years on the Shansabanis, they are still deemed less than consequential in
modern Western-language historiography, remaining in the shadows of
their more famous contemporaries such as the Ghaznavids of Afghanistan
and the Saljuqs of Iran, Central Asia, and Anatolia (c. ah 431–590/1040–
1194 ce).12 This striking divergence between Western and “indigenous
epistemological values” – to modify Thomas de Bruijn’s phrase for our
immediate purposes13 – certainly reiterates the fractured and contingent
nature of regional “knowledges” at any given time. Indeed, this divergence
leaves open the long and probably futile task of integrating “knowledges”
into anything near a singular and shared Knowledge, even with the new

Figure I.10 Mosque known as “Arhai Din-ka Jhonpra,” founded, c. 1199 CE. Ajmer (Rajasthan, India), interior:
prayer area, view of columns and ceilings. © Didier Tais 2011.
14/09/21 5:36 PM
10 Iran to India
Figure I.11 Mosque of Kaman, Rajasthan, India, founded c. 1195. Exterior: eastern
façade with main entrance. © Photograph Alka Patel 2011.
technologies and modes of communication constantly becoming available.

On a more humble (and more optimistic) level, however, the present
volumes’ treatment of the Shansabanis at least mitigates their historio-
graphical displacement. At the same time, I believe the subject presents
important opportunities to craft much-needed methodological alterna-
tives for approaching the historical pasts of South and Central Asia.
In their magisterial work titled Empires in World History, J. Burbank
and F. Cooper proposed a capacious conceptualization of empires, viz.
“large political units, expansionist or with a memory of power extended
over space … that maintain distinction and hierarchy as they incorporate
new people.” Such a working definition not only leaves room for a great
variety of politico-economic hierarchical structures – that is, empires –
through time, it also respects their staggering complexity. Burbank and
Cooper’s definition easily encompasses such dissimilar entities as the
Roman and Mongol empires, bookending two soft ends of a spectrum
of imperialisms: Rome, with its “notion of a single, superior imperial
civilization open in principle to those who could learn its ways”; versus the
Mongols, who “learned their statecraft from both Eurasian and Chinese
sources … sheltered Buddhism, Confucianism, Christianity, Daoism, and
Islam; [and] where diversity was treated as both normal and useful.”14
But a singular term for such a vast array of human organizations
certainly belies the virtually infinite possibilities for their origins; their
subsequent expansions and contractions; and their internal political, eco-
nomic, social, and other operations. Empires were never instantaneous,

Figure I.12 Congregational Mosque, Herat, Afghanistan, c. 1200 CE, exterior: southeastern portal (blocked).
© Photograph Alka Patel 2011.
14/09/21 5:36 PM
12 Iran to India
Figure I.13 Congregational Mosque, Herat, interior: western iwan, vault.

monolithic, or even unidirectional. Not only has empire-building always

occupied varying lengths of historical time, the regimes in question and
their elites were often perceived differently, depending on the time and
place within said empire – again evoking the poet Sana’i’s homily of the

Introduction 13
Figure I.14 Tomb of Ghiyath al-Din Muhammad ibn Sam, Herat (addorsed to
northern perimeter of Congregational Mosque), prior to destruction in 1944. From
Glatzer 1980: fig. 12 (photograph R. Stuckert, 1942).
elephant and the blind men of Ghur. Moreover, the methods for achieving
empire were ad hoc and determined by specific circumstances, rather than
prescribed beforehand.
Primarily – though not exclusively – relying on the Shansabanis’ archi-
tectural patronage, the present work aims to reconstitute the proverbial
elephant into a perceivable whole. It may appear contradictory, then, to
pursue this aim in two volumes rather than one, but I believe this approach
in itself presents an important opening. As the coming pages will show,
the corpus of Shansabani foundations is datable by means of inscription,
textual reference, and/or stylistic comparanda as falling within c. the early
1100s to c. 1215 ce – a corpus herein documented largely first-hand, ana-
lyzed, and also expanded. Such an architectural group permits a concen-
trated gaze on approximately eighty–ninety years of the mobilization of
regional labor structures and the formation of new political hierarchies
toward empire-building – an intimate entrée that is rare in the study of

14 Iran to India
Figure I.15 Tomb of Ghiyath al-Din Muhammad ibn Sam, Herat, reconstructed
c. 1940s. © Photograph Alka Patel 2011.
any region or polity of nearly one thousand years before the present. But
while Shansabani ascendancy was temporally brief, it was geographically
and culturally vast, spanning the nomad–urban continuum of central
Afghanistan (described below), along with the cosmopolitan emporia of
Khurasan to the west; and to the east, the multi-confessional landscapes
of the Indus’s alluvia, and reaching beyond to the equally variegated north
Indian duab.
With the Shansabanis’ victories in northern India during the 1190s, a
qualitative temporal and spatial divide seemed to crystallize between their
western and eastern territorial presences, especially in light of the varyingly
constituted histories and historiographies of these regions – and indeed
those of the eastern Islamic world and South Asia overall. And yet, these
regions were ultimately intertwined. The two-volume framework here
allows for the analysis of their material and discursive archives separately,
but necessarily in dialogue, each with the merited analytical focus. This
volume (Volume I) concentrates on the areas now encompassed within
modern Afghanistan and Pakistan, spanning the early twelfth century
through c. 1190 ce. It thereby prepares the way for Volume II, which
examines the Shansabanis’ extremely consequential expansion into the
Indo-Gangetic plains (c. 1192–1205 ce), and also traces the westward rever-
berations of the Shansabanis’ politico-military successes. Volume II thus
closes the cultural–geographical circle, creating a fuller picture of their
short-lived but historically significant imperium, in other words an archi-
tectural biography.

Introduction 15
Such a close examination of approximately eight–nine decades in the

lifespan of the Shansabanis’ confederate empire presents manifold addi-
tional avenues for inquiry. Among the foremost aims for the project at
hand, however, is the untethering of scholarship on South and Central
Asia’s history of the second millennium ce from the apparently irresistible
pull of a magnetic pole originating in pre-modern historiography, namely,
the drive to write emphatically Islamic histories.
In the Arabic and Persian histories of the Islamic world dating roughly
from the thirteenth through eighteenth centuries, we can justifiably
observe that among their authors’ pre-eminent drives was the adumbra-
tion of not only a “divinely ordained paradigm of the rise and fall of
states,” but specifically an Islamically ordained one.15 Such a drive was
plausibly even more pronounced on the perceived peripheries of this phys-
ical and moral world, as would have been the case in the environs of South
and Central Asia. Arabic- and Persian-language authors writing about the
region tended to treat the Shansabanis as one lock in the ever-lengthening,
multi-stranded braid of Muslim dynasts. For some, the first lock of this
braid would have been the Umayyad incursions of the eighth century,
though the Ghaznavids might have shared this initiatory status particu-
larly with regard to northern India.16
The continued presence of specifically Islamic dynastic lineages in
South Asia was not lost on these historical authors (collectively): the
above-mentioned braid of events continued its intricate and sinuous pro-
gress with the Shansabanis and then the so-called Delhi Sultanate (usually
dated c. 1211–c. 1525 ce), which also engendered the creation of eventually
independent states governed by confessionally Muslim elites through-
out the Deccan. Ultimately, the advent of the Mughals and what were
ostensibly the latter’s successor states of the eighteenth century brought
this Islamically ordained timeline into the very present of some of these
authors.17 In hindsight, then, even while the authors’ presentation of the
sequence of events was not strictly incorrect, their understanding of the
Shansabanis’ place in near-exclusively Islamic historical time arguably
cleaved to the central purpose of writing an Islamic history. Furthermore,
the composition of Islamic pasts was very much in keeping with the histor-
ical authors’ moral universe(s), and the epistemological and literary con-
ventions guiding their intellectual labors (see also Chapter 1, this volume).
Today, it is safe to say that the universes – moral and otherwise –
providing the impetuses for scholarship are different, as are the episte-
mological and intellectual conventions guiding present-day authors. The
seemingly inescapable urge to identify historical actors and processes
within an Islamic, divinely ordained “history” does not have to be preva-
lent, and the need to legitimize a royal patron’s rightful place in Islamic
time and space is now obsolete. And yet – jarringly – the disciplinary
specialties of “Islamic” history, art and architecture, archaeology, etc. have

16 Iran to India
only grown in appeal and practitioners. And, as if echoing the aforemen-

tioned historical authors’ collective urge to place their monarchs within
Islamically ordained spacetime, scholarly works on South Asia’s historical
periods still elide various political formations – even with vastly different
court cultures and socio-cultural geographies – into single monographs
principally because of the Islamic confessionalism of these polities’ elites.18
In the face of these countervailing cross-currents, it is difficult – if not
close to impossible – to “essay … non-sectarian histories of conquest and
conflict,” and also cultural transformation.19
The ensuing volumes essay20 just such a non-sectarian retelling. They
embrace the possibility for new understandings of South and Central
Asia’s history of the second millennium ce, offered by liberating the
point of departure from an inherited urge to write an Islamic dynastic
history. Certainly, the chapters that follow focus on the Shansabanis as a
kinship-based elite, or dynasty (Figure I.3), and they trace their trajectory
from unknown origins toward a confederacy of royal lineages exercising
prerogatives broadly encompassed within notions of Perso-Islamic king-
ship – what could be termed a history. The work does not, however, take it
for granted that this kinship-based group were always elites, nor that their
relationship with Islamic confessionalism was a seamless one. The fissures
thereby opened afford the opportunity – thus far too little utilized in the
scholarship on Central and South Asia – to examine which “Islam” was
favored by these eventual elites, and why (see esp. Chapters 2–4).
Of course, prior modern works have undertaken close analyses of his-
torical dynasties of the Persianate world, perhaps most relevant being
the monographs on the Ghaznavids, considered the predecessors of the
Shansabanis in the regions now within modern Afghanistan through India
(cf. Chapters 1 and 7). Prior to becoming prized ghulams in the service of
the Samanids (c. 820–1000 ce), the Ghaznavid dynasty’s early founders
Alptigin (d. ah 352/963 ce) and his own ghulam Sabuktigin (c. 940s–999
ce) – father of the famous Mahmud (r. ah 389–421/999–1030 ce) – both
came from obscure origins in the vast steppe lands of Turkic and other pop-
ulations, roughly spanning modern Uzbekistan and Kazakhstan through
Mongolia.21 The broad outlines of Sabuktigin’s truly eventful life indicate
that his integration into and eventual rise within the Samanid house were
undeniably underpinned by a series of dialogical encounters with more
than one aspect and type of “Islam.”22 Indeed, his rule on behalf of his
Samanid overlords in Zabulistan-Zamindawar (southern Afghanistan;
Figures I.1–I.2) and surrounding areas also included the deployment of
selective “Islams” in the interests of statecraft (cf. Chapter 3, notes).
But the existing studies on the Ghaznavids, emerging from the last
century of scholarship, treat them principally as Perso-Islamic kings who
constructed an empire, once extending from Rayy in the center of the
Iranian Plateau to Lahore in the Panjab, and possibly beyond (see esp.

Introduction 17
Chapters 5 and 6). By contrast, rather than assuming largely unprob-

lematic transactions between a singular notion of kingship and a similar,
broadly conceived Islam in Central–South Asia during the early second
millennium ce, the present close study of the Shansabanis takes full
advantage of the available evidence to examine the identifiable impetuses
behind their actuations of kingship and specific processes of Islamization
(see Chapters 1 and 2). We focus on a slice of time, viz. the long twelfth
century, during which the Shansabanis emerged from unknown origins
into historicity.
In Sana’i’s homily, the role of the blind men of Ghur would be played
by the scholars from several disciplines who have undertaken analy-
ses touching upon the Shansabanis during a little more than the last
half-century. These specialists have included historians, archaeologists
and art-architectural historians – among the last of whom I would count
myself. As specialists of “archives” encapsulating time periods and geo-
graphical areas, we understandably embrace our disciplinary locations.
These allow us to stay abreast of the continually changing methodologies
afforded by each of the aforementioned fields, and the sheer technological
advances that also engender new approaches.23 In fact, the present mon-
ographic project relies on the architectural patronage of the Shansabanis
throughout modern Afghanistan, Pakistan, and India as a primary source
to elucidate the architectural cultures in which they participated, and which
they shaped in turn, as explicated below. But, I believe that in the case
of the Shansabanis and the northwestern reaches of South Asia (modern
Afghanistan, Pakistan, India) during the twelfth century, all too rigid and
exclusive disciplinary loci have led to misperceptions of the “elephant” as
a whole.24
Objects and texts dating to the early second millennium ce constitute
important evidentiary corpora, elucidating what is generally a less accessi-
ble time and place of Eurasia’s past. The textual sources on eleventh- and
twelfth-century Afghanistan through India, which are the primary focus
for historians, have been extensively analyzed and interpreted over the
last century and more. Similarly, the Shansabanis’ architectural and other
material traces throughout their imperial expanse have attracted the exper-
tise of art-architectural historians (cf. esp. Chapters 2–5 and Volume II).
In fact, these bodies of evidence have been juxtaposed and even deployed
in tandem, thereby combining the information from more than one part
of the proverbial elephant. As a result, F. B. Flood summarized thus the
modern perception of the Shansabanis, which has remained more or less
unchanged over the last sixty years25:
Although ultimately ephemeral (and thus largely ignored by historians and

art historians alike) … [the Ghurid polity’s] conjoining [of] northern India
with large swaths of eastern Iran and Afghanistan for the first time [sic] …

18 Iran to India
created the conditions for mobility between these contiguous realms on a pre-
viously unimaginable scale. The expansion of the Shansabanid lands to include
vast areas that had traditionally fallen outside the dar al-Islam presented the
parvenu Ghurids and their agents with a double imperative: to project their
authority and legitimize their position within the dar al-Islam and to provide
effective governance in the newly acquired Indian territories that had formerly
stood without it.
While this description of the Shansabanis is – again – not strictly

incorrect, it accrues principally from two parts of Sana’i’s elephant, viz.
those elucidated by the rather rigidly circumscribed disciplinary loci of
history and art-architectural history. Fortunately, the Minaret of Jam
Archaeological Project (MJAP), directed by Dr. David Thomas, has
recently brought to light a third pachydermal part, making apprehension
of the whole “beast” that much more attainable: rigorous archaeological
surveys and excavations by MJAP specialists during the mid-2000s in
central Afghanistan’s Ghur province, at the site of Jam-Firuzkuh (Figure
I.2) – now generally believed to be the sard-sir encampment or summer
“capital” of one branch of the Shansabani confederacy (cf. Chapters 2–4) –
provide us with invaluable data of the material conditions of the lives
of the Shansabani elites and at least some of their cohabitants at the
locality.26
Making sense of the whole range of seemingly unrelated or only
circumstantial data for the Shansabanis, as well as the existing lacunae
therein, is not unlike the sensation of following the elephant’s trunk (for
example), only to discover an entirely new and seemingly dissimilar yet
ultimately connected part of an organic terrain. It has necessitated forays
into the additional discipline of anthropology, and therein the study of
nomadic pastoralists and other transhumant populations, often referred to
as “barbarians.”27 The principal hypothesis animating the present project
is that, during the first few decades of the twelfth century, the Shansabanis
experienced an ancestral shift from being more nomadic and/or seasonally
transhumant (“barbarians”) to more sedentist power-brokers, not neces-
sarily reaching either end of the nomad–urban continuum.28 Recently,
Patricia Crossley further clarified that:
[n]omads can be designated more concretely as pastoralists – people at some

moment in time depending primarily upon the care and processing of animals
for a living, but always living and trading within a broader context of agricul-
tural, urban economies.29
Following Crossley, then, for convenience I use the less cumbersome

“nomadism” and “transhumance” and their adjectival derivatives to encom-
pass the full range of nomadic pastoralism and seasonal transhumance.
As for the Shansabanis and the consequential changes to their lifeways,

Introduction 19
these undoubtedly took place in response to a variety of social, economic,

political, and perhaps environmental factors (see Chapters 2 and 3). But
as is the case with all historical processes, in the end the surviving evi-
dence allows for the perception of only some of these factors.30 Again,
Burbank and Cooper have provided perhaps the most succinct description
of the accidental convergence of circumstances that eventually coalesce as
imperialism:
In conditions of wide access to resources and simple technology, small

advantages – larger family size, better access to irrigation or trade routes, good
luck, ambitious and skillful rulers – can lead to domination of one group over
another, setting in motion the creation of tribal dynasties and kingships.31
I have found such an apprehension of the Shansabanis’ transformation

to be invaluable for making fuller sense of many, previously opaque or
misunderstood aspects of their trajectories, both within Afghanistan and
beyond.
It is also worth noting that, unlike other nomadic/transhumant groups
who attained transregional politico-economic power – most relevant
would be the contemporaneous Sajluqs (cf. esp. Chapters 1 and 3) – the
Shansabanis’ shift from textually unrecorded nomadism and/or transhu-
mance to the elite status celebrated in panegyrics, and ultimate imperial
demise, are all documentable within approximately eighty to ninety years.
Their nomadic/transhumant beginnings, then, would not have been a
distant memory evoked in the nostalgia of elders; some form of these
lifeways was conceivably still within the lived experience of the scions
of the three principal Shansabani lineages at Firuzkuh, Bamiyan, and
eventually Ghazna (Chapters 3–5). Thus, the Shansabanis’ pre- and early
imperial history – the focus of Volume I – is directly relevant to our
investigation of their transformation into transregional builders of empire
in Volume II.
Bennet Bronson’s well-known and tantalizingly titled essay, “The Role
of Barbarians in the Fall of States” (1988), is equally useful in examin-
ing the reverse, state-building process. Defining “barbarians” essentially
as members of “a political unit … that is not itself a state,” the author
summarized thus their situation:
If rich barbarians successfully resist both their state-organized and unorganized

but poorer neighbors, it is probably because they have set up a state them-
selves. Barbarians who remain barbarians are therefore poor.32
Following Bronson’s inimitable directness, many of the Ghuri “barbar-

ians” remained poor, while the Shansabanis obviously did not. All the
more reason to be cognizant of a distinction, pertinent especially as of
the mid-twelfth century: the historical inhabitants of the region of Ghur

20 Iran to India
were and are logically referred to with the toponymic adjective Ghuri;
however, a specific kinship-based clan emerged from among them, iden-
tified by historical sources as “the house of Shansab” or the Shansabanis.33
As discussed in Chapter 2 this distinction becomes even more significant
over time, as the majority of the Shansabanis’ fellow Ghuris most likely
did not undergo a similar shift in lifeways and economies from varieties of
nomadism or transhumance to some degree and type of sedentism (or to
empire-building dominance).
To be sure, historical writers such as Minhaj al-Din Siraj Juzjani
(ah 589–658/1193–1260 ce) – our principal interlocutor throughout this
work (cf. Chapter 1) – referred to the Shansabanis also as “the maliks
of Ghur,” and the appellation has understandably continued in modern
Persian- and European-language scholarship.34 However, in addition to a
question of historical discernibility and greater accuracy regarding their
politico-military ascendancy as of the mid-twelfth century, using the clan
name of “Shansabani” for the protagonists of our story is also historio-
graphically meaningful: To refer to the Shansabanis as “Ghurids” might
reify the misinterpretation of their historical impact as limited to their
region of origin and – in Barry Flood’s term – ephemeral (see above).
But as Flood himself rather contradictorily stated, their activities rever-
berated on “a previously unimaginable scale,” so that distinguishing the
Shansabanis as such from their fellow Ghuris seems only fitting.35
The present Volume’s examination of the Shansabanis also provides
an entrée into central Afghanistan’s nomad–urban continuum, the more
nomadic aspects of which have historically received less than their merited
attention in a wide range of disciplines from history to archaeology and
beyond (see Chapter 1). Certainly, the relative paucity of surviving mate-
rial culture from nomadic societies – or, an alternative characterization
could be that methods adequate to recovering these material traces are
still coming into focus – has led to fewer studies on the histories of these
populations.36
The real significance of nomadic groups is increasingly evident, not
least because they interacted with more sedentist and urban populations
to varying degrees, often as “peripatetic traders” supplying desirable com-
modities and/or labor, and exchanging or purchasing goods within urban
and village contexts. Indeed, the seasonal migrations of nomads occa-
sioned regular interactions with the settlements, villages, and towns along
their routes of transhumance.37 Nomads are therefore extremely important
in fully grasping the economic and cultural layers of historical towns and
cities, which all together shaped the landscape often over centuries.38 The
ensuing analysis attempts to contribute to a recuperation of more of the
socio-cultural history of central Afghanistan, relevant both to apprehend-
ing the region’s historical periods and also to understanding it more fully
in modern times.39

Introduction 21
Nevertheless, as alluded to above, the nature of the material remains

of past nomadic/transhumant societies is only one of a series of challenges
surrounding their retrievability. Given that most of the textual sources
emerged from urban and even courtly ambits (mentioned previously), it is
unsurprising that non-sedentist populations appeared in these texts only
in passing, the reports occasioned by their interactions with the hegemonic
power in a particular region (cf. Chapter 1). Moreover, many texts’ raison
d’être was to chronicle a regional or transregional power and enable its
longevity in collective memories, so that the official reports of this power’s
interactions with nomadic and transhumant groups rarely depicted them
as anything but “barbarians” (cf. Chapter 1). The result, then, is a virtual
lacuna of source materials on historical nomads that are contemporaneous
and, if not sympathetic, then at least interested in the vast spectrum of
nomadic existences.40
This historiographical blind spot can be somewhat ameliorated by
studies of modern nomadic societies. While these studies by no means
provide direct substitutes for the ideal source materials just described, they
can offer documentation of phenomena in very specific contexts, provid-
ing interpretive simulacra that aid us in devising plausible interpretations
of the surviving traces of past nomadic and transhumant groups. Bearing
in mind Richard Tapper’s41 salutary cautions against the ahistorical use of
studies on modern nomads to understand their historical antecedents, two
broad parameters for their consultation are in order: first, relying on works
treating modern Afghanistan and Iran for comparanda permits geograph-
ical consistency, underscored by an active awareness of the alterations in
climate and natural resources within these regions’ various landscapes over
the last one thousand years.42 Second, rather than a narrow, one-to-one
application of modern data to past events, judicious juxtaposition tends to
be less anachronistic, and more open-ended in the range of meanings that
come into view. Within these parameters, the picture that emerges of the
Shansabanis frames a shift from pre-imperial nomadism/transhumance
among their fellow Ghuri countrymen, to a regional and then transre-
gional dominance, albeit with continuation of some aspects of their earlier
lifeways (see esp. Chapters 2 and 3).
This project’s focus on an obscure, transhumant/nomadic group of
the past, and the elites emerging from it, might appear to be an unlikely
subject for an art-architectural historian. These populations certainly pro-
duced portable objects such as metalwork and textiles, made possible by
their regular purchase or barter of raw materials in towns and settlements,
where in turn they might sell or trade the fruits of their labors.43 But they
had no need to construct monumental architecture, generally not possess-
ing the requisite skills themselves, nor patronizing specialized workers to
build for them (cf. Chapter 2).
A premise of the current work is that Perso-Islamic kingship during the

22 Iran to India
period of the Shansabanis’ emergence as a regional power in the twelfth

century required cultural patronage, including architecture, wherein so
many of the religious, intellectual, and juridical supports of rule were
housed (see esp. Chapter 1). But rather than viewing architecture as
solely an instantaneous phenomenon, being the monumental result of
elite patronage, a perception of the built form within its larger nexus of
processes of facture, varying uses, and shifting receptions all together aid
in accessing a wider cross-section of its historical significances. Indeed,
precisely because command and mobilization of skilled labor was requisite
in the creation of built works, these serve as diagnostic points of reference
for the emergence of new politico-economic elites – that is, previously
unknown groups who came to possess the political and material capital
required to have buildings constructed for them. Akin to Sana’i’s elephant,
then, discrete aspects of the built work hold immense potential for differ-
ing yet complementary insights into its milieu, and the groups of people
who interacted with it throughout its existence (see, e.g., Chapter 2).
But, as modern investigators attempting to access a past, how might
we understand the architectural “product” that historical patrons were
underwriting, either through command or compensation? What were the
range of standards that circumscribed its outcome? I propose the concept
of architectural culture as an accommodating heuristic device to apprehend
the collective span of the built works analyzed herein.
The writings of J. Sourdel-Thomine (e.g., 1960, inter alia) and
M. Meister (esp. 1993, 2011) regarding “style,” “idiom,” and “mode” have
initiated the difficult but essential process of coming to grips with coalesc-
ing patterns in the history of material–cultural production. These scholars
had – very responsibly – applied their observations to their respective
regions of expertise, Sourdel-Thomine distilling her long experience in
the eastern Islamic world (greater Khurasan), and Meister his equally
profound engagement with the building traditions spanning the Indus’s
alluvia through the north Indian plains. Collectively, these authors’ writ-
ings are not merely relevant to the regions encompassed within the emerg-
ing Shansabani elites’ annexations, they have formed the bases for my
conception of architectural culture as a complex of constitutive mecha-
nisms shaping their respective built environments, in turn allowing us to
address the nature of the Shansabanis’ participation within them.
This Volume and the next argue that the Shansabanis’ territories –
eventually extending into north India – encompassed three distinct and
equally significant architectural cultures: modern central through western
Afghanistan, interconnected with historical Khurasan; the Indus’ alluvia;
and the Indo-Gangetic plains.44 The building groups located in these
regions did not simply look different from each other: They were in
fact convergent indices of several factors, including indigenous natural
resources; materials of construction; architectural traditions and their

Introduction 23
praxes; and established labor structures or hierarchies. All of these factors

together flexibly constituted the architectural culture operative in a perme-
ably delineated locus, which served as the principal resource for architec-
tural patronage among elite and non-elite echelons alike.
It must be emphasized that an architectural culture at a moment in
time was far from monolithic, and the concept should be capacious in
its synchronous application. Here, I find Holly Edwards’s45 notion of
“spectrum piety” applicable within an architectural context: the virtually
infinite historical iterations of an architectural culture spanned a wide
spectrum, encompassing elite patronage, which is typically thought to set
far-reaching, cosmopolitan standards. But not all structures in a given time
and space were commissioned by elites or built by the same artisans. Since
non-elite patronage also manifested the architectural culture in question,
all its iterations should be viewed with parity, as points on the spectrum
spanning elite or royal, cosmopolitan patronage, through non-elite or
vernacular commissions. Rather than one being of lesser or greater value,
all instantiations of an identifiable architectural culture serve as valid and
equally valuable historical informants.
As such, an architectural culture would rarely evince drastic and sudden
changes with shifts in political régimes, for example, or even the introduc-
tion of new religious practices.46 Really, new architectural cultures took
root within a region only with the influx of skilled laborers and the availa-
bility of the required building materials, both in terms of natural resources
(e.g., stone deposits, earth usable for bricks) and “industrial” infrastructure
(quarries, kilns, and fuel, etc.). In fact, multiple architectural cultures could
co-exist within a region,47 as the mobility of skilled laborers is a known phe-
nomenon,48 with the demands for their skills depending on the vagaries of
their patrons’ preferences. But while a regional, politico-economic elite by
definition commandeered material resources and labor toward their own
requirements, political shifts did not abruptly or even necessarily interrupt
the inertia of a region’s architectural culture(s), or the ways in which
buildings were constructed or decorated. At the same time, it is also true
that all unmechanized human practices change, more or less perceptibly,
over time. The relationship between architectural patronage and the built
form, then, cannot always be characterized in general terms, but largely on
an individual and ad hoc basis.
The tomb of Salar Khalil (also known as the ziyarat of Baba Khatim)
(Figures I.16 and I.17) in Afghanistan’s Juzjan Province, about 40 km west
of Balkh, and the tomb locally attributed to Shahzada Shaikh Husain
near Bust (Figures I.18 and I.19) both serve as discrete and informa-
tive exemplars of an architectural culture in general, and in particular as
manifestations of greater Khurasan’s architectural culture, which occu-
pies the majority of this Volume.49 Along with minars, and mosque and
madrasa complexes, the “domed cube” configuration of these tombs was

24 Iran to India
Figure I.16 Tomb of Salar Khalil, a.k.a. ziyarat Baba Khatim (west of Balkh),
Juzjan Province, Afghanistan. From Sourdel-Thomine 1971: fig. XI.
a long-standing and ingrained building practice throughout Central Asia

by the latter half of the twelfth century, utilized in both monumental
commemorations of royalty and the nobility, and in smaller-scale struc-
tures, as is the case in these examples.50 Salar Khalil’s otherwise extensive
epigraphic program recorded no date, while Shaikh Husain’s exterior
inscriptional frieze was too eroded to glean the content. Although the
subject of some debate (cf. below), both tombs have been dated to the
twelfth–thirteenth centuries.51 These structures are amply useful for our
purposes, as they illustrate the inclusive notion of an identifiable archi-
tectural culture and the variability within it, even at the non-royal ends
of its expressive spectrum. They also demonstrate the heuristic value of
identifying an architectural culture, which both shaped and was shaped
by the patronage of elites and others alike.
While both tombs adhere to the overall format of a domed space, they
diverge in their respective funerary chambers, one being quadrangular
(Juzjan) and the other octagonal (Bust). Moreover, the Bust tomb was
largely open to the elements, more akin to a pavilion sheltering a grave,
while the Juzjan tomb was a closed “cube” with only the entrance aperture.
The most noticeable difference between them, however, is the amount of
decoration and the variation in the respective iconographies: the exteriors

Introduction 25
Figure I.17 Tomb of Salar Khalil, Juzjan Province, interior. From

Sourdel-Thomine 1971: fig. XV.
and interiors of the Bust structure (Figures I.18 and I.19) carried decora-
tions primarily consisting of baked bricks placed in horizontal or vertical
clusters, and terracotta plugs deployed in a remarkable variety of ornament,
ranging from textured surfaces; geometric patterns, including a plethora of
medallions; to the angular Kufic frieze running along the exterior of the
dome’s octagonal base. Meanwhile, the Juzjan tomb (Figures I.16 and I.17)
seemed to act as the stage for an elaborate epigraphic program of variations
on the Kufic script against equally diverse backgrounds. Both the differ-
ences and commonalities in these tomb structures – located at two ends of
the early Shansabanis’ core imperial regions in Afghanistan – are worthy
of note, as together they personify two of the manifold iterations within a
region’s architectural culture.

26 Iran to India
Figure I.18 Tomb of Shahzada Sheikh Husain, near Bust, Helmand Province,
Afghanistan, c. 1200 CE. From Crane 1979: fig. 5.
The debate alluded to above, specifically surrounding the tomb of Salar

Khalil, is significant as it illustrates the heuristic value of an architec-
tural culture. The debate has centered on whether the structure merits
a Ghaznavid or “Ghurid” attribution, based largely on the form and, to
a lesser extent, the content of its epigraphy.52 Indeed, any purely dynas-
tic attribution for the structure in the twelfth century would be further
complicated by the extremely intricate history of the Balkh region overall,
which was a point of contestation among the Khwarazm-shahs, the
Firuzkuh and/or Bamiyan branches of the Shansabanis, and the western
Qarakhanids as Qarakhitay vassals (see Chapter 1; Figure 1.1).53
Salar Khalil is in fact remarkable due to the predominance of doxolo-
gies, and Qur’anic verses and references throughout its inscriptions; the
use of an unmistakably hard-edged Kufic script with geometric terminals
in the baked-brick exterior frieze (Figure I.15); and on its interior walls
(Figure I.16), the juxtaposition of meandering calligraphic bands with
incised stucco surfaces.54 Based on these considerations, I concur with
Sourdel-Thomine (1971) that the structure was most likely built during the
latter half of the twelfth century, during the waning of Ghaznavid power
in Afghanistan and these sultans’ effective retrenchment to their eastern
holdings around Lahore. The question of a definitive dynastic affiliation
for the structure is not only difficult to answer, it may in fact be largely
irrelevant, as the fractious political circumstances in the Balkh region
during the later twelfth century seemed to exercise no visible impact on its
form and decoration.

Introduction 27
A structure such as the

tomb of Salar Khalil not only
exquisitely personifies the
many factors contributing
to its own process of becom-
ing – that is, its architectural
culture – it also demon-
strates the impediment that
an overdetermined reliance
on dynastic labeling can
create55: rather than facili-
tating entry into historically
germane processes, the quest
for a top-down dynastic
appellation obscures our
apprehension of changes in
architectural culture that
were incrementally effected,
at least to some degree by the
dialogue between shifts in
patronage and the artisans’
responses to them. It was just
such a dialogue that the newly
ascendant Shansabanis had to
master if they were to achieve
dominance both within their
lands of origin and beyond. Figure I.19 Tomb of Shahzada Sheikh Husain, near Bust,
And it is this fundamental interior. From Crane 1979: fig. 8.
process that buildings help us
to understand, whether among well-heeled Persianized elites grasping for
an empire, or politico-cultural novices emerging from the nomad–urban
continuum, thrust forward by what were essentially historical accidents
such as “better access to … trade routes, good luck … [and] ambitious
and skillful rulers.”
* * *
Before embarking in earnest on this Volume’s examination of the emer-
gence of the Shansabanis on the eastern fringes of Khurasan during the
mid-twelfth century, and their eastward progress to the Indus’s alluvia
shortly thereafter, the difficulty of fieldwork in large swaths of Afghanistan –
including the central region of Ghur itself – must be acknowledged. Given
the inaccessibility of many of the early Shansabanis’ core areas of activity,
projects such as mine perforce rely on historical and other documentation,
which has come from a wide variety of sources.

28 Iran to India
Figure I.20 “The fortress and citadel of Ghazni (Afghanistan) and the two Minars.
Tribesmen with camels, horses and pack bullocks in foreground,” James Atkinson
(1780–1852), watercolor, 1839. © Courtesy of the British Library.
Figure I.21 “Ruins of Old Kandahar Citadel,” albumen print, c. 1880–81,

Benjamin Simpson. © Courtesy of the Getty Research Institute, Los Angeles.

Introduction 29
Being a photogenic land of truly magnificent landscapes, architecture,

and colorful people, Afghanistan has arrested the picturesque imagination
at least since the First Anglo-Afghan War (1839–1841) (Figure I.19), con-
tinuing to do so into the Second Anglo-Afghan War (1879–1881) (Figure
I.20) with the great popularization of photographic technology as a tool
to “visit” distant and exotic lands without the inconvenience and risk of
actual travel. This fascination with Afghanistan seemed to persist during
the first eight decades of the twentieth century, as archaeologists and
other scholars undertook more systematic surveys and studies. These latter
works included the invaluable archive of photographs left by the intrepid
Josephine Powell (1919–2007) (Figure I.21).56 Throughout more than a
century of depiction, Afghanistan’s lofty ruins and nomadic populations
appear unchanging (compare Figures I.21 and I.22), always furnishing
aesthetically evocative subjects begging to be Orientalized. I myself could
not help but capture the wide-open plains edged by mountains outside the
Figure I.22 Shahr-i Gholghola, general view, citadel (eleventh–thirteenth century),

with women in foreground. © Photograph Josephine Powell c. 1960. Harvard Fine
Arts Library, Special Collections.

Figure I.23 Herat-Ghuriyan highway, Herat Province, Afghanistan. © Photograph Alka Patel 2011.
14/09/21 5:37 PM
Introduction 31
western city of Herat (Figure I.23), appearing as natural proscenia upon

which great histories unfolded.
But such modern fetishism actually distracts from the long-standing
reality of these eastern parts of the Islamic-Persianate world. The regions
within modern Afghanistan were collectively a busy place, being the stra-
tegic areas for control of much of Eurasia, as realized by the Sasanians,
various Hunnic groups such as the Hephthalites, and the Western Turks –
to name only a few who built and consolidated empires starting and some-
times ending there. Moreover, the picturesque but artificially empty vistas
occlude the multifarious iterations of the above-mentioned nomad–urban
continuum prevailing throughout the vast majority of the Iranian cul-
tural world, including Afghanistan, well into the nineteenth century. The
protagonists in forging imperial formations were not solely the regions’
magnificent landscapes; they combined with enterprising populations,
which spanned the nomad–urban continuum, to link the Iranian and
Indic cultural spheres. Overall, it is to a more complete apprehension of
this dynamism that the ensuing project is dedicated.
Notes
1. Translated by J. Stephenson 1911: 13. On this British officer in the Indian

Medical Service and his early study of Sana’i’s extremely complex Hadiqat
(including the work’s codicological history), see de Bruijn 1983: 122; and
Allegranzi 2019a, vol. I: 29–30.
2. According to Lewis (1995: 120), it was Sana’i who first introduced the Buddhist
parable of the blind men and the elephant into the literature of the Persianate
world, a parable thereafter incorporated into the works of al-Ghazali (ah
450–505/1058–1111 ce), Jalal al-Din Rumi (d. ah 671/1273 ce), and others. Cf.
also de Bruijn, “Sana’i” EIr (2012); but cf. also note infra.
3. Although “Udana” characterized a variety of Buddhists texts, principally in
Pali, here the reference is to the third book of the Khuddakanikaya, which
contained eighty stories and utterances of the Buddha. The utterances were
mostly in verse, accompanied by prose accounts of the circumstances prompt-
ing them. The Buddhist elephant parable was predictably situated within the
sacred geography of early Buddhism, so that the king referred to in the text
ruled at Shravasti (Pal. Savatthi), where he summoned “all the men born
blind in Savatthi” to describe an elephant. During the historical Buddha’s
lifetime (mid-first millennium ce), Shravasti was the capital of the powerful
Koshala kingdom, located in the foothills of the Himalayas of modern Nepal.
Cf. Buswell and Lopez 2014: 42, 383, 435, 443, 932; see also the English trans-
lation of the Pali Udana by D. M. Strong (1902: 93–96).
4. Thomas de Bruijn, among the principal scholars of Sana’i, described Hadiqat
as “one of the most problematic works of Persian literature” for Western
scholars, though in the Persian ecumene it has enjoyed a favored status over
the centuries, providing “a promising source for the study of indigenous

32 Iran to India
literary values” (de Bruijn 1983: 119). Given Hadiqat’s homiletic genre and
overall structure – didactic poetry paired with illustrative prose anecdotes –
further juxtaposition with the Pali Buddhist canon may aid in dispelling the
current confusion around its complex form. Cf. de Bruijn 1983, 119–163; de
Bruijn, “Sana’i” EIr (2012); and Lewis 1995, 112–120. Indeed, such an inter-
disciplinary juxtaposition could also be fruitful for questions regarding the
specific Buddhist texts that circulated throughout historical Bactria, and these
texts’ audiences and receptions.
5. Twelfth-century inhabitants of the regions within modern Afghanistan
were surrounded by remnants of Buddhist monasteries and other structures.
According to Melikian-Chirvani (1974: esp. 34ff., 64–65), certain “stereo-
types” with identifiably Buddhist literary origins had also filtered into Persian
poetry. While Lewis (1995: 120) concluded that Sana’i was the first Persian
poet to have introduced the elephant parable into Persian poetry and litera-
ture (see note supra), the specific mechanisms by which Sana’i accessed the
contents of Buddhist literature – was he versed in Pali, and were manuscripts
available in Ghazna or accessible in the region? – are still to be determined.
Alternatively, the poet might have made use of an already translated Persian
version, in which case he would not have been the first Persian poet to inte-
grate the parable in his œuvre.
6. I borrow the phrasing from Peacock’s (2010: 3) important study on the early
Saljuqs.
7. Aside from Ghaznavid-issued coinage – including the well-known Kufic-
Sharada dirhams and figural jitals, the latter continuing in the Panjab for
several centuries – no surviving architectural remains farther east in the
duab region are known. Cf. R. Tye and M. Tye 1995: 43 and nos. 86, 89,
91–94.
8. Cf., e.g., Rosenfield 1967/1993: 41–54; also Ball, Bordeaux et al. 2019: 345. The
convergence of Greek, Iranian, and Indian “symbols of power and victory”
in Kushana iconography underscored the regular traffic of people from all
of Eurasia’s major cultural regions within Kushana-affiliated territories.
Additionally, among the legacies of the Kushana period would be the coa-
lescence of the anthropomorphic iconographies of Buddhism, which was to
have enduring effects on the religio-cultural lives of peoples across Eurasia for
centuries to come. Cf. Pons, “Kushan Dynasty ix. Art of the Kushans,” EIr
(2016).
9. I have found the recent engagements by Green (2019) and Spooner (2019)
with the “Persianate” – with some modifications (cf. Chapter 1) – as a heuristic
useful for delineating the process of attaining kingship that the Shansabanis
undertook.
10. See, e.g., Patel 2010 for an overview of textual sources indicating seventh–
eighth-century presence of Muslim communities in South Asia; cf. also Asif
2016: 13. The Shansabanis’ north Indian activity is the principal subject of
Volume II.
11. The origins of this catch-all rubric have been attributed to early twentieth-
century scholarship on India (cf. Patel 2018: esp. 60) and the singularity of
the term “Sultanate” in the face of the multiplicity of these polities and their

Introduction 33
cultural productions has also been interrogated (ibid., 60–67; see also Patel
2006: 9–11). Additionally, see infra in main text and notes.
12. The Persian-language monographs – two published in Afghanistan and two
in Iran – were recently analyzed against the backdrop of these areas’ respec-
tive nationalisms, emerging over the twentieth century. While the “extra-
intellectual” impetuses of these works’ authors were brought to the fore, “the
cognizance of [such] factors on scholarship … in western-language works”
must also be borne in mind (Patel 2017: 144 ff.). Further, see Thomas (2018:
9 and passim) for a discussion of “the Ghurid interlude,” coined by Bosworth
(1968b: 166).
13. See de Bruijn 1983, cited supra in notes (my emphasis).
14. Burbank and Cooper 2010: 4, 8, 12–13, 15, 17–18.
15. Meisami 1993: 268. See also Hardy 1960: 113–118; Sarkar 1977: 28; Nizami
1983: 48–49; Siddiqui 2010, esp. 2–8. See also Auer 2012: 3ff.
16. For the early eighth-century Umayyad incursions into Sindh and their usages
in colonial and later historiography, see Inden 2000: esp. 17 and notes. Also,
while Asif (2016: 2ff.) discussed Chach-nama’s (c. 1226 ce) description of
these events as the dubiously grand “beginnings” of Islam in South Asia, the
campaigns of Mahmud of Ghazna into northern India have also commanded
attention as originary in public and intellectual cultures over the last millen-
nium (cf. Anooshahr 2009: 1–3).
17. See, e.g., Sen (2017) for analysis of Muhammad Ghulam Husain Tabatabai’s
Sair al-Mutakhirin (1779–80). For the earlier period of the sixteenth century,
Anooshahr (2009: 23–26) intriguingly posited that the future emperor Babur’s
motivations for (and his perception of) his own campaigns into India were
deeply colored by his thorough familiarity with Mahmud Ghaznavi’s raids into
the north Indian duab as gleaned from the works of ‘Utbi and Juzjani. These
interpretations of events and processes form a striking contrast with Bronson’s
(1988: 208–209) contemporary historicist view of northern India as “… one of
the world’s great civilized traditions [that was] within raiding range of unusu-
ally effective barbarians” (my emphasis) – cf. infra in main text and notes.
18. For a relatively recent example, see Flood’s Objects of Translation (2009),
whose timespan encompassed the ninth through the early thirteenth cen-
turies, and the historical geographies of western Afghanistan through the
Panjab and the Himalayan foothills, as well as the north Indian plains. The
author stated: “… the centuries covered by this book have occupied center
stage in colonial and nationalist constructions of a past that has been cast as
a perpetual confrontation between Muslim invaders and Hindu resisters, a
Manichean dyad that has structured and constrained the history of the region
for almost a millennium” (ibid., 2) – essentially reiterating the tautological
inertia the current work is attempting to interrupt.
19. Cf. Amin 2002: 30, again with slight modification in the form of adducing
the historical process of cultural transformation, for the purposes of this
Introduction and the broader aims of these volumes.
20. Applied much in the same sense as Amin (cited supra), i.e., “to attempt, to
try to do, effect, accomplish, or make (anything difficult),” in use as of the
seventeenth century. Cf. OED, “essay.”

34 Iran to India
21. This approximate regional span is taken from Cahen et al., “Ghuzz”, EI2
(2012). These authors admittedly concentrated on the western Oghüz and
other Turkic groups, as these were mentioned in Arabic- and Persian-
language texts, often through first-hand accounts of the merchants and pros-
elytizers encountering them (see Chapter 2). The Oghüz’s eastern brethren,
the Toküz–Oghüz, remained more remote and lesser known.
22. Nazim (1933: 610–614, trans. 622–628) first published Sabuktigin’s putatively
autobiographical Pand-nama, which was to be found only as copied in a
fourteenth-century text. It briefly mentioned Sabuktigin’s boyhood among
the Turkic Barskhan clan, his abduction and sale at the slave market of
Nakhshab (modern Qarshi in Uzbekistan), and ultimately the Samanid
ghulam Alptigin’s purchase of him in Bukhara. Sabuktigin described the
worship of stone idols among some Turkic tribes and his own skepticism of
the practice, indicating his pre-destined conversion to Islam. Sabuktigin’s
origins also included the historiographically requisite Sasanian connections,
cf. Bosworth 1973: 61. See also Barthold 1968: 261–262. Such a biographical
trajectory was common among the slaves who rose to the status of ghulams, as
also indicated by the parallel story of Qutb al-Din Aibek – cf. Juzjani, vol. I,
1963: 415–416 (trans. vol. I, 1881: 512–514).
23. See, e.g., Wilkinson 2016: 3. A caveat is required here: satellite data usually col-
lected for US and other super-power intelligence agencies have concentrated
on more populated areas, “river valleys and along roads, leaving blank spaces
especially in mountain and desert areas” (cf. Hammer et al. 2017: 5). Thus, at
times even this otherwise invaluable resource can inadvertently reinforce his-
torical lacunae in knowledge rather than remedy them. Nevertheless, the areas
whence satellite data have recently been analyzed include the northern arc of
Balkh and the Balkhab River, as well as the historically less explored southern
reaches of Afghanistan around the Helmand River and Sistan. Satellite images
of the latter, for example, have revealed at least 119 caravanserais built approx-
imately every 20 km. Dated to the late sixteenth–early seventeenth centuries
and considered to be part of a “centrally sponsored effort,” these way-stations
connected the Safavid (1501–1732) and Mughal (1525–1858) realms in uninter-
rupted trade and travel routes, altering the view that the Safavid empire had
entered a politico-economic decline by this period. Cf. Lawler 2017: 1364;
Franklin and Boak 2019. For a recent study on hydrological development
and change in the Balkh oasis during the ninth–thirteenth centuries – based
on satellite data, in large part – see Wordsworth 2018. Other satellite data
have reconfirmed earlier archaeological work in regions such as the vicinity of
Qandahar, where ceramic finds imported from Safavid Iran were the most sub-
stantive corpus unearthed by excavations (Crowe 1996: 314, 320). Provided all
dating is accurate, these rediscoveries underscore the centrality of Afghanistan
for the Indic and Iranian cultural worlds during the early modern period. See
esp. Chapter 4 for the use of Google Earth and other satellite imagery toward
the apprehension of historical data not previously gleaned from architectural
complexes and their surrounding landscapes.
24. A perfect case in point would be the mid-twentieth century’s strict divi-
sion of intellectual labor among disciplines, amply demonstrated in works

Introduction 35
of the eminent historian C. E. Bosworth (1928–2015), and those of the

equally renowned and coeval anthropologist Fredrik Barth (1928–2016):
Despite shared research geographies, the purely textual approach of the his-
torian remained singularly unaffected by the on-the-ground, albeit modern
findings of the anthropologist (and vice versa). Cf. infra in main text and
Chapters 2–4.
25. Flood 2009: 91. This recent overall assessment of the Shansabanis has remained
largely consistent beginning with Bosworth’s scholarship of the 1960s, and
extending through that of H. Edwards in the 1990s–2010s.
26. One of the major results of MJAP’s efforts is, of course, Dr. Thomas’ 2018
monograph, resulting from the only twenty-first-century archaeological work
at Jam-Firuzkuh, from which the present book has benefited immensely. For
the archaeo-botanical and -zoological analyses undertaken by MJAP, see, e.g.,
Thomas et al. 2006; and for ceramics esp. Gascoigne 2010 (inter alia).
27. Cf. Vasjutin 2003: 52; Mahmud 2009: 251–252; Elliott 2013: 5; Peacock 2015:
3ff. and notes. The idea of “barbarian” has, of course, derived from the seem-
ingly parasitic raids on caravans and settlements by unindustrialized nomads.
Bronson (1988) pointed out that “barbarians” possessed varying degrees of
literacy and technological sophistication, and were considered “barbarians”
simply because they lay outside a state’s apparatus in terms of mobility and
modes of production (i.e., varied and small-scale agricultural practices,
“cottage” industries, sporadic and less taxable trading). Cf. also infra in main
text and Chapter 1.
28. Cf. Thomas and Gascoigne 2016: 171–172. This premise is at variance with
most previous studies, which assumed that “the population of the region [of
Ghur] was largely sedentary, and had apparently been so since ancient times”
(Habib 1992: 2). See also Bosworth 1961: 118, 133 n. 53. For the updated general
characterization of the Shansabanis and their fellow Ghuris here, I rely on
Thomas’ monograph (2018, esp. pp. 3–10ff.). Cf. also Chapter 2.
29. Crossley 2019: xviii. Cf. also the work of the Collaborative Research Centre,
“Difference and Integration,” a joint project at the universities of Halle-
Wittenberg and Leipzig: Büssow et al. (2011: 3) summarized the group’s over-
arching “assumption that [nomads] follow rational choices no less than their
sedentary counterparts … [for] example … preoccupation with the securing
of pasture rather than an innate (and therefore irrational) lust for plunder
[urging] the Turkic ‘tribes’ who had come to Anatolia in the eleventh century
[to] head for the Caucasus.”
30. Tracing the emergence of empires from nomadic or otherwise transhumant
origins can be as fraught with preconceptions and generalizations as any
other inquiry into large and multi-faceted socio-historical processes, under-
scoring the need for ongoing work to finesse more specific explanations.
While anthropologists and historians have tended to favor some theories of
state formation and the impetuses behind it, it is understood that no singular
factor can explain the phenomenon among sedentist, nomadic, or in-between
populations. For other overviews of favored theories of nomadic imperial
formation, see Kradin 2003; also Crossley 2019.
31. Burbank and Cooper 2010: 9.

36 Iran to India
32. Bronson 1988: 200. Cf. also Vasjutin 2003: 51.

33. The poet NiÕami ‘Aru∂i (fl. twelfth century) used the nomenclature in his
Chahar Maqala ([trans. Browne] 1921: 8). Juzjani’s explication of the myth-
ical progenitor Shansab (or Shanasb) was itself derived from a lost gene-
alogical work by Fakhr al-Din Mubarak-shah al-Marwarruzi, a nadim of
Sultan Ghiyath al-Din – not to be confused with Fakhr-i Mudabbir (cf. also
Chapters 1 and 6). See Juzjani, vol. I, 1963: 318–319 (trans. vol. I, 1881: 302);
also Siddiqui 2010: 18.
34. Juzjani, vol. I, 1963: e.g., 318 (trans. vol. I, 1881: 300). See esp. O’Neal (2013:
22–30) for an informative overview of the historiography, beginning with the
European “discovery” of the Shansabanis as Persian historical texts under-
went translation in the late seventeenth century, through studies dating to
the early to mid-twentieth century. Cf. Patel 2017 for modern scholarship in
Persian.
35. The seemingly paradoxical tendencies of the long-term effects of dynastic
lineages and their ephemerality appears to have been endemic to nomadic
empires, as observed of the transregional nomadic empire(s) par excellence
(Burbank and Cooper 2010: 93 and passim): “The empires established by
Mongols were not long-lived, at least when compared to Rome or Byzantium.
What makes Mongols count in world history are the connections they made
across Eurasia and the imperial technologies they adapted, transformed, and
passed on to later polities.”
36. Among the initial recalibrations of methodology for the archaeology of
nomadic societies has been Cribb’s work (1991), emphasizing the indices of
mobile groups’ encampments within areas of study. Thomas (2018: 17, 50)
effectively deployed the resulting concept of an archipelagic landscape, coined
by Adam T. Smith, in which “urban centers [are] small islands of data linked
by ephemeral networks in a sea of general obscurity” [!]. See also Stark 2005;
Stark 2006a, 2006b; and Thomas and Gascoigne 2016: 173–176.
37. Cf. supra in main text; also Barfield 1981: 102–103; and Crossley 2019: 15–16
38. Notwithstanding the outdated generalization of “Middle Eastern Tribalism,”
particularly apropos here are Ernest Gellner’s (1983: 422; ibid., 445) early
observations summarizing the contributions to a volume principally on
tribal–pastoralist groups in Iran and Afghanistan: “These tribal societies are
accustomed to a level of technology, in their agricultural, pastoral, military
and domestic equipment, which seems to pre-suppose centres of artisan pro-
duction and trade, in other words towns …” Given the nuancing over time of
nomad–sedentist interactions, however, Elliott (2013: 6) proposed modifying
Gellner’s (1983: 447) earlier characterization of “the [nomadic] tribe [as] the
anti-state” as actually a relationship that is both symbiotic and oppositional,
already implied by Bronson (1988).
39. Certainly, the Afghan Boundary Commission reports of the mid- to late
nineteenth century make constant reference to nomadic communities, a
characteristic also photographed by Donald Wilber around Herat in the
1930s, and by the indefatigable Josephine Powell throughout Ghur during
the 1960s (cf. infra in main text). These images harmonize with A. Kohzad’s
“tour reports” (see esp. 1951a; 1952; 1954a; 1954b; also discussed in Chapter

Introduction 37
2), as well as late-twentieth-century studies of nomadic groups in central

Afghanistan, e.g., Glatzer and Casimir 1983 – the last study being especially
pertinent as the same scholars documented and published the magnificent
Shansabani madrasa of Shah-i Mashhad (Figure I.1) (Badghis province; dis-
cussed in Chapter 4) in the late 1960s while following the migration patterns
of Pashtun nomads. Indeed, Najimi (2015: 147) noted the seasonal presence
of the Achakzay nomads in the madrasa’s vicinity in recent years, continuing
the migration patterns noted by Glatzer (1983). See also Thomas 2018: 26.
40. Comparable challenges exist for other nomadic elites, most notably the
Saljuqs: cf., e.g., Durand-Guédy 2010: 70–71; and esp. Durand-Guédy 2013a
for the interpretation of textual sources. Also Peacock 2010: 6–14; Peacock,
2015: 2–19; and Chapter 1.
41. Tapper 1997: 18ff.
42. See Paul 2018b: 316–318 for just such a juxtaposition of modern and historical
nomadism in the vicinity of the Balkh oasis. Bosworth (1961: 118) noted,
e.g., that: “After the Mongol invasions the pastoral element among the pop-
ulation [of Ghur] increased; whereas [earlier] the Tajik inhabitants … had
been primarily agriculturalists”; but he provided no specific evidence for the
statement, seeming to rely on the assumption that the nomadic Mongols
would have introduced or strengthened ways of life akin to their own. But
long- and short-term changes in climate have had obvious impacts on society
as a whole, and should be one of the primary considerations in differenti-
ating historical and contemporary realities. While comparative data from
Europe on the “Medieval Warm Period” (c. 700–1300 ce) might provide
points of departure, extrapolating larger global trends would be conjectural
at best. Nonetheless, Thomas (2018: 28) summarized that: “[t]he decima-
tion of Afghanistan’s natural flora and fauna in recent decades … means
that modern data cannot be viewed as representative of the Early Islamic
period” (eighth–thirteenth centuries). A possible case of drastic change in
the landscape – whether due to environmental factors, shifts in riverbeds or
irrigation practices, or a combination of all three – is the region of Sistan:
here Klaus Fischer, one of this region’s most ardent and thorough inves-
tigators, noted an “unusual extent of ruins” compared with the sparseness
of the modern population (Fischer 1969: 337). But a further observation,
viz. “the majority of watch towers, fortresses, rural estates (rustaq), ‘open
towns’ (cities without walls) situated in the plain and fortified cities existed
from the beginning of the 12th to the end of the 15th cent. A.D.,” was only
scantily evidenced by surface findings, thus Sistan’s architectural chronology
has remained inconclusive. Cf. Fischer et al. vol. II, 1974: XII; Genito 2014;
see also Chapters 1 and 2.
43. As highlighted by Mahmud (2009: 251–252), reiterating A. A. Kohzad’s
(e.g.,1951a, 1952, 1954a/b) observations during the course of his travels
throughout Ghur in the 1950s. See also Thomas and Gascoigne 2016.
44. Sourdel-Thomine (1960: 280; my trans.) avowed that: “… [t]he art flour-
ishing in Afghanistan during the sixth/twelfth century deserves above all
the qualification of a post-Saljuqid Khurasani art developed from the tradi-
tions of the region …” For the northwestern Indo-Gangetic region, see esp.

38 Iran to India
Meister 1993; and for what he termed the Gandhara-Nagara style of West
Panjab and Khyber-Pakhtunkhwa (modern Pakistan) see Meister 2010; 2011
(and Chapter 6). For a full description of the three architectural cultures
within Shansabani domains, see esp. Patel 2007/2011.
45. Edwards 2015: 71ff.
46. Cf. Patel 2004.
47. Sourdel-Thomine (1960: 280) declared with surprising rigidity that “post-
Saljuqid Khurasani art [in Afghanistan] developed without undergoing the
least innovation worthy of interest, for example upon contact with India”
– an ahistorical understanding of stylistic purity that will be addressed in
Volume II.
48. See Volume II for the explication of just such a process: during the last decade
of the twelfth century and the first two decades of the thirteenth, artisans
specially skilled in the building materials, concepts, and iconographies of
northwestern India – in modern times eastern Sindh, Gujarat, and Rajasthan
– made their way westward, along the corridors of mobility newly invigorated
by the Ghazna Shansabanis’ eastward expansion.
49. For the Juzjan structure, see Melikien-Chirvani 1968; Sourdel-Thomine 1971.
For Bust, see Crane 1979. For overall bibliography, cf. Ball 2019: Nos. 149,
439.
50. See Grabar 1966 for early Islamic funerary structures, and esp. Pugachenkova
1978 for commemorative architecture in the environs of Balkh, where she
documented a long trajectory of the architectural configuration through at
least the fifteenth century. See also Hillenbrand 2000: 138; and Chapters 2, 4
and 6.
51. While the Juzjan tomb provided the forms and content of its extensive
epigraphic program – discussed infra in main text (cf. also Appendix,
Afghanistan I) – the one at Bust required other analytical tools: Crane (1979:
245–246) based his dating of the tomb on comparisons with similar decorative
techniques in Iran and Central Asia, as well as the independent stelae found
within the structure, which had a date range from the twelfth to thirteenth
centuries. See also Sourdel-Thomine 1956; and Volume II.
52. The first publication of the tomb by A. Melikien-Chirvani (1968) attributed
it to Ghaznavid presence in the region, while Sourdel-Thomine (1971) argued
for Shansabani patronage.
53. See Paul 2018b: 323ff. for Balkh’s shifting political fortunes during the twelfth
century; for the numismatic evidence, cf. Schwarz 2002: e.g. Nos. 811, 824,
834; and the more synthetic study by O’Neal (2020). A similar debate has
surrounded the tomb of al-Hakim al-Tirmizi (northwest of modern Tirmiz,
Uzbekistan), with an epigraphic program noting only an ambiguous patron
but no date, so that its construction and/or renovation has been attributed
either to the eleventh century, possibly by a local western Qarakhanid ruler,
or (most recently) to the mid-twelfth century: see Allegranzi 2020: 112ff.
54. Chapters 4 and 5 discuss the distinguishing stylistic features of Shansabani-
period epigraphy, as well as the preponderance of Arabic-language and
Qur’anic content along with the possible reasons for the latter. Suffice it
to say here that the Qur’anic verses and references gracing the structure

Introduction 39
(Appendix, Afghanistan I: 3–5) were not unusual in the overall corpus of

Islamic architectural inscriptions: e.g., Qur’an III: 18–19 (Al ‘Imran) was
also part of epigraphic programs at Chisht and Ghazna, and throughout
the eastern Islamic lands since at least the eleventh century (cf. Blair 1992: 9,
52, 56, 137, 153); meanwhile, the famous Throne verse (II: 255–256) was also
present since at least the early eleventh century at Tirmiz (Blair 1992: 101),
and at Chisht and Lashkari Bazar. Cf. also Chapters 4 and 5, and Appendix,
Afghanistan III: 5, VIII: 4.
55. Blair (1985: 84) took up the application of dynastic appellations to art-
architectural styles in her analysis of the Zuzan madrasa (ah 615/1218 ce),
patronized by the local malik in the service of the Khwarazm-shah ‘Ala’
al-Din Muhammad Takish (r. ah 596–617/1200–1220 ce).
56. Even during the 1970s prior to the Soviet invasion of 1979 (and for scholars
based within Afghanistan itself – e.g., N. Dupree 1977), Powell’s photographs
were regularly utilized both in popular guidebooks and more scholarly works
as among the most thorough and meticulously documented archives of the
more remote areas of the country. Powell’s work was incredibly prescient,
making possible the continuing research on this global region into the present
day.

CHAPTER 1
Kingly Trajectories
T he chronological scope of this two-volume project, namely ah 490s–

610s, overlapping with the twelfth century of the Common Era, has
been determined by the Shansabanis’ appearance in historical records,
and their imperial rise, expansion and dénouement along these tempo-
ral lines. The current volume focuses on the Shansabanis’ early history;
their initial expansions beyond Ghur-Zamindawar (see maps, pp. xv and
xvi); and their fracturing into separate, geographically distinct lineages
(see Genealogy, p. xvii) – a confederate empire – all taking place during
c. ah 510–580s/1120s–1180s ce (cf. Chapters 3–5). The emphasis of the present
volume, then, is on relevant events dating to approximately seven decades
of the twelfth century in what is modern Afghanistan. Such a specific tem-
poral and spatial focus might imply that a historical overview of the early
Shansabanis would be largely a question of recapitulating the information
in textual sources and existing scholarship. In fact, this is far from the case,
for reasons that are historical as much as they are historiographical.
Although obviously never devoid of “history” per se, nomadic and
transhumant groups rarely encountered the need to record in written
form the events constituting their past. The challenges of recuperating oral
histories are well known, moreover, especially as orally transmitted narra-
tives underwent “translation” upon becoming written.1 Concomitantly,
the court-centered nature of most imperial chronicles and other textual
sources meant that nomads and other transhumant populations received
only passing references, their appearances in texts occasioned primarily
by interaction with the hegemonic “state” in a particular region.2 Indeed,
the elision – if not erasure – by contemporaneous textual sources of the
nomadic past even of the great Saljuqs (tenth–twelfth centuries), and a
parallel silence regarding the continuation of non-sedentist mores in their
imperial phase,3 are all amplified in the case of the geographically remote
Shansabanis. But, in addition to being a question of general historical
import, a view of the Shansabanis during the late eleventh–early twelfth
centuries – even in broad strokes – furnishes important data to bear in
mind as the later rulers’ trajectory broadened into an imperial arc, encom-
passing the valleys of the Indus and ultimately the plains of northern India
(cf. Chapter 6 and Volume II).

Kingly Trajectories 41
The absence of major trading centers within Ghur meant little to report
for the early Arabic-language geographers,4 or other authors such as the
anonymous writer of the tenth-century Hudud al-‘Alam (c. 982 ce) –
which betrays the proclivities and aims of these authors (and their modern
heirs) rather than the historical importance of the region. In a similar
vein during the eleventh century, the Ghuris rarely arrested the attention
of Ghaznavid-period chroniclers such as al-‘Utbi (c. 961–1036/1040 ce)
and Baihaqi (ah 385–469/995–1077 ce), and when they did, it was not in
flattering terms: they were recipients of punitive action for their banditry
along caravan routes. The three well-known Ghaznavid expeditions into
Ghur – two by Sultan Mahmud (r. ah 389–421/999–1030) to the fortified
settlement of Ahangaran on the Hari Rud (cf. Chapter 2) and Khwabin in
southern Ghur, and a third by his son Mas‘ud (eventually Sultan Mas‘ud
I, r. ah 421–432/1030–1041) to the fortress at Jurwas (map, p. xvi) – were
motivated both by the cessation of previously stipulated tribute to the
Ghaznavid treasury, and by the Ghuris’ disruption of caravan traffic.5
Considering the practical impossibility of entire caravans traversing the
difficult terrain of the region, the Ghuris’ brigandage likely took place on
the major routes radiating from Herat and essentially skirting Ghur itself,
either proceeding northeast toward Balkh and eventually India through
the Khyber and other Gandharan passes; or headed southeastward
toward Tiginabad/Old Qandahar, Baluchistan, Sindh, and ultimately the
Indo-Gangetic plains.6 Although these reports provided few specific facts,
such as exactly where the bandit raids took place or indeed the commod-
ities the raiders sought in waylaying the caravans,7 they did indicate in
broad terms that at least segments of Ghur’s inhabitants practiced some
form of banditry, likely as a supplement to other modes of production.8
Furthermore, the lack of sustained passage of outside populations through
Ghur surely also impacted its linguistic culture, as the dialect of Persian
spoken here – or perhaps another language altogether – required Prince
Mas‘ud to rely on translators during his campaign.9 The prevailing under-
standing of the region up to the twelfth century has largely followed the
eminent C. E. Bosworth’s (1928–2015) characterization: it had “no towns
of note, but only agricultural settlements and – [the] most typical feature
of the landscape – fortified places and towers (qasr, qala, hisar, kushk)”10
marked a politically fragmented region, whose largely pastoral and belli-
cose inhabitants “practis[ed] agriculture and banditry [and] specialized in
the production of weapons and war equipment.”11
While the overall picture of historical Ghur and Ghuris is further
enriched throughout this volume (see esp. Chapter 2), the ongoing schol-
arly efforts to re-integrate nomadic and transhumant populations into
fuller apprehensions of regional histories and economies should be noted
here. For example, M. S. Mahmud’s Persian-language monograph on the
Shansabanis12 has initiated a much-needed re-examination of the evidence

42 Iran to India
pertinent to Ghur. Citing the plentiful natural resources and mines of

the region, as well as its well-known commodity of slaves, this author
proposed that Ghur must have been linked with transregional commercial
networks despite the absence of such references in the historical texts.13
After all, Ghur’s oft-cited mountainous landscape surely presented more
difficulty for outsiders seeking access into the region, than for those long
familiar with the terrain as they transported their goods out.
Thereafter in the twelfth century, the macroscopic perspectives of univer-
sal histories such as Ya‘qut’s (ah 575–626/1179–1229 ce) Mu‘jam al-buldan
and Ibn al-Athir’s (d. ah 630/1233 ce) often cited al-Kamil fi al-Tarikh
both mentioned Ghur and its inhabitants only infrequently. Even though
they were by now more properly the ruling Shansabanis – identified thus
by writers such as Nizami ‘Aruzi, and Juzjani (see below) – they neverthe-
less continued to play minor roles in the eyes of the universal historians.
As was the case in texts of various genres from the previous centuries, the
Shansabanis made appearances principally when they intersected with
the larger imperial formations of the Saljuqs and the Khwarazm-shahs.
The expanded historical gazes and raisons d’être of these universal histories,
then, seemed to alienate further this corner of the eastern Persianate lands
and its inhabitants from the flow of the larger Islamic world.14
But at the same time, by the mid-twelfth century and the Shansabanis’
initial ascendancy as a regional politico-military presence, the influx of
unknown or unfamiliar power-seekers – warlords becoming warrior kings,
to adapt terms used by M. A. Asif and A. Anooshahr15 – and indeed their
resounding successes within the Persianate world, were all undeniable and
even well-known phenomena. So much so, that the invention of royal
genealogies for newcomers was an integral labor performed by the schol-
arly literati whom these new courts patronized (cf. below). These invented
genealogies created “a teleology of Islamic monarchies, [adhering to the]
logic of historiographic expectations,” which obviously occluded the rising
groups’ historical origins.16
Such a set of impetuses and intentions inhered in Tabaqat-i nasiri
(completed before 1260), a universal history within the eponymous
tabaqa genre of Persian historical writing. The work’s author was the
jurist and accomplished courtier Minhaj al-Din Siraj Juzjani (ah 589–658/
1193–1260 ce) who, moreover, had been brought up within the Shansabani
royal household at Firuzkuh. Despite Tabaqat’s ostensibly universal and
trans-temporal purview, the disproportionately long sections (Ar.-Pers.
tabaqa) on the Shansabanis revealed Juzjani’s true focus. Following in the
established tradition of royal genealogies, the author bestowed upon the
Shansabanis the expected lineage springing from pre-Islamic mythology,
whose principal repository was the Shah-nama and the literary production
it engendered.17 Admittedly, the requisite genealogical “origin myth” created
for the Shansabanis was unusual, descending through the figure of Azhd

Zahhak – a “repugnant figure in the Iranian lands farther to the west …

[but commanding] popularity in the region of Ghazna and Zabulistan.”18
Indeed, Juzjani dutifully provided a surfeit of origin stories for the
obscure transhumant clan who would come to be recorded in his and
others’ chronicles as the Shansabanis. Addressing an array of the histo-
riographical registers of his time, the author also granted a seemingly
unassailable Islamic pedigree to his patrons, claiming their conversion to
Islam by none other than the khalifa ‘Ali himself during the first century of
the Hijra.19 Additionally, Chapter 2 posits that Juzjani couched a possible
reference to their rural and probably nomadic-transhumant origins and
the beginnings of their transformation to elite status: A story in Tabaqat
centered on the Shansabanis’ mythical progenitor Amir Banji and the
friendly Jewish merchant who became his ally and advisor comes across as
the parable-like narrative of upward politico-economic mobility. As apoc-
ryphal as the story undoubtedly is, we are nonetheless tempted to credit
Juzjani with an uncanny prescience, as if he had laid the crumbs along the
evidentiary trail sought by the modern investigator.
Tabaqat ensued from Juzjani’s own family history and recollections. The
author was close to and invested in the Shansabanis’ activities in Khurasan
(modern Afghanistan and contiguous areas; map, p. xv), as a result produc-
ing a virtually unique account of the early dynastic history of his patrons.20
Additionally, Tabaqat drew from another, equally first-hand work that
is unfortunately no longer extant: a versified genealogy by Fakhr al-Din
Mubarak-shah al-Marwarruzi (d. ah 601/1205 ce).21 This Persian-Arabic
poet and writer on astronomy had dedicated the work to Sultan ‘Ala’
al-Din Husain Jahan-suz (r. ah 544–556/1150–1161 ce), and thereafter had
continued in royal service, being mentioned as a nadim in the retinue of
Sultan Ghiyath al-Din Muhammad ibn Sam (r. ah 558–599/1163–1203 ce).
Juzjani apparently consulted the genealogy in the possession of Mah Malik,
the daughter of Sultan Ghiyath al-Din residing at Firuzkuh.22
Ghur’s main economies of nomadism–transhumance rather than
commerce, and the region’s general difficulty of access meant that the
aforementioned universal histories, such as Ibn al-Athir’s Kamil, often
gathered information that was second-hand – though it was not nec-
essarily unreliable (cf. below). Moreover, rather than re-litigating the
factuality of reported events among historical sources – already under-
taken by others23 – Juzjani and his Tabaqat act as the primary textual
interlocutors throughout this volume. Not only is Tabaqat the sole
surviving work essentially devoted to the Shansabanis, it also furnishes
the most fleshed out account of their early trajectory prior to the east-
ward expansions of the Ghazna branch – albeit according to its author’s
priorities (cf. below and Chapters 2–4). Juzjani’s own professional and
geographical movements curiously recapitulated those of his Shansabani
patrons: he faithfully served them throughout much of his life, also

44 Iran to India
from Lahore with the addition of their eastern territories, until their
demise in the second decade of the thirteenth century. The Mongol
campaigns of the 1220s ce compelled Juzjani to take refuge in Delhi,
where he completed Tabaqat sometime before his death in 1260 ce.24
But beyond this ontogenetic resonance between author and subject,
Tabaqat also presents rare insights: throughout the work, subtle but dis-
cernible indices emerge of Juzjani grappling with the tensions between
the “historiographic expectations” of twelfth-century Persian historical
writing, and his own vested interests in crafting his patrons’ memorial
power for posterity. What we learn from these tensions, and the fissures
between expectations and “reality,” will be discussed later in the chapter
and throughout this volume.
Attaining Kingship: Perso-Islamic Cultures of Power
The historical overview of the early Shansabanis in the latter part of this
chapter reveals the lacunae that persist in our knowledge of their activi-
ties, even after the marshalling of evidence from material–cultural sources
such as numismatics and architectural patronage. But while the minutiae
of the Shansabanis’ rise to regional and then transregional ascendancy
may not be fully perceivable, tracing even its broad contours presents the
opportunity to reflect on the Persianate ecumene specifically during the
twelfth century. This temporally focused analytical gaze on the admittedly
longue durée historical phenomenon of the Persianate permits, I believe, an
enrichment of its more punctual manifestations.
At variance with recent reconsiderations of the concept of the
“Persianate” – introduced by M. Hodgson in the last quarter of the twen-
tieth century – I propose that, for the place and time of our investigation,
“Perso-Islamic” is a more apt descriptive term for the elite cultures dis-
cussed forthwith.25 Furthermore, rather than conceiving of these cultures
of power, or prestige cultures, as inaccessible except to those already within
their ambit – that is, elites begetting elites – we must concede that courts
throughout central and western Eurasia were often established by new-
comers to the Persianate cultural world. What made this possible?
In attempting to answer such a broad question, I do not aim to define
per se Perso-Islamic cultures of power, as they differed in their details at
each court according to myriad historical contingencies (see, e.g., Chapter
3). Instead, it is more productive to adopt an inverted approach: what
aspects of these prestige cultures facilitated access to kingship for ambi-
tious interlopers such as the Shansabanis? Taking a cross-section of the
twelfth century, and particularly the intersecting imperial formations of
the Ghaznavids, the Saljuqs, and, eventually, the Shansabanis in what is
now Afghanistan (map, p. xv), we can propose some tentative answers

to the question of how unknown and patently non-elite groups attained

kingship during this historical moment.26
Perso-Islamic prestige cultures should be understood as lexicons for rule,
or collective “playbooks,” which, when mastered and enacted, were tanta-
mount to actuations of kingship in central–western Eurasia. It was this vast
complex of elite cultures that effectively allowed populations far-removed
from cultural capitals both in distance and lifestyles – especially nomadic
and/or transhumant groups – to become power-brokers and even rulers. A
group of unknown Ghuris exemplified such a process, eventually emerg-
ing as politico-military elites styling themselves as Shansabanis. Their
emergence and other, similar developments throughout the Persianate
world underscore – counterintuitively – the essentially egalitarian nature
of Perso-Islamic court culture: a recognizable language or code of behavior
that could enable kingship for pretty much anyone with the acumen to
master it, and the resources to implement it.
What, then, were the aspects of twelfth-century Perso-Islamic cultures
of power, or lexicons for rule, that up-and-coming groups could command
in order to actuate kingship? Rather than static constituent elements or
“ingredients” comprising these lexicons, it is more fruitful to conceive of
three, overlapping and even fungible spheres of activity, with appreciably
wide ranges of possible expressions: one sphere was Islamic confession-
alism; a second, the large-scale patronage of cultural production in both
intellectual as well as material forms; and third – though by no means
least important – was mobility. It might seem unnecessary to point out
that the last two spheres of activity were not peculiar to Perso-Islamic
kingship, but they deserve emphasis as they were reiterative of Islamic
confessionalism, and were arguably fundamental to the ascendancy of the
principal powers of the twelfth-century Persianate world. All three of these
overlapping spheres of activity left historical traces in the landscapes the
Shansabanis inherited, thus providing them with aspirational models for
attaining kingship. The rising Shansabanis came to act within these three
spheres – elucidated further below – in tandem with accruing regional and
then transregional economic, political, and military primacy.
Further, rather than conceiving of Perso-Islamic kingship as unbounded
and infinitely reproducible, positing its approximate geographical extent is
useful as a counterpoint. The Qarakhitay (c. 1125–1218 ce), a nomadic
power who occupied the Balasaghun region (modern Kyrgyzstan) as of
the 1130s ce, seemed to personify the eastern borders of Perso-Islamic
kingship’s currency. Although being in intimate contact with their vassals
the Qarakhanids (c. 998–1213 ce) and other subordinate powers such as the
Khwarazm-shahs (c. 995–1231 ce) – who espoused Perso-Islamic kingly
affiliations – these polities’ Qarakhitay overlords emphatically did not.
Rather than converting to Islam, the Khitay rulers retained Buddhist–
animist and Christian affiliations, creating a composite kingly culture of

46 Iran to India
“Khitan tribal elements [and] assimilated Chinese features” even as they

ruled over a majority Muslim population comprising vassals and citizenry.
In the eyes of their subject populations in Transoxiana and those farther
east, Qarakhitay legitimacy was associated with the “high civilization” of
China. Even though the region comprising this civilization was a concept
often devoid of geographical and demographic accuracy, it behooved the
Qarakhitay to maintain their eastern loci of aspirations: their imperial
domains lay at the intersection of Persianate and Sinicizing ecumenes,
throughout which such a projection of kingly personas was legible, effec-
tive, and accepted.27
During the twelfth century, the Yamini-Ghaznavids’ (c. 998–1186 ce)
instability attracted politico-military rivals into the pivotal region now
bounded by modern Afghanistan. For the Ghaznavids, their core holdings
in Zabulistan, al-Rukkhaj, and Zamindawar (map, p. xv) had collectively
served as a springboard both toward the west, deep into Khurasan, as
well as the east, to the Indus valleys and the north Indian plains. Thus,
these core holdings were both strategically and commercially desirable
(cf. Chapter 5). They were also where the Shansabanis were emerging
as a local power and carving out the beginnings of their own imperium,
embroiled in a long and fractious relationship with the Ghaznavids (see
below and Chapters 2 and 3). The Shansabanis’ rise in the fourth through
eighth decades of the twelfth century also instigated confrontations with
the Saljuqs (tenth–mid-twelfth centuries) as the latter sought suzerainty
over the smaller polities on their eastern frontiers (cf. below), and the
Khwarazm-shahs (twelfth–thirteenth centuries) and Qarakhanids (tenth–
early thirteenth centuries), who were also attempting territorial expan-
sion.28 Afghanistan, then, was the canvas upon which the trajectories of
the Persianate ecumene’s major empires – which eventually included the
Shansabanis – intertwined throughout the twelfth century.29
Although on quite different geographical scales, the spectacular debut
of Saljuq power in the Persianate world during the eleventh century,
and the meteoric rise of the Shansabanis in the twelfth, evince fasci-
nating historical and historiographical parallels. Most significantly, the
Saljuqs and Shansabanis were interlopers, entering from “outside” the
elite Perso-Islamic prestige cultures they subsequently deployed toward
kingship. The Saljuqs’ Turkic ethnie and earliest associations with the
territories of the tenth-century Khazars (in modern west Kazakhstan) were
not their only exogenous characteristics; the principles of socio-political
organization and the many cultural traditions arising from their nomadic
life-world also rendered them largely “other” to the Persianate regions they
came to dominate.30
The Shansabanis’ ethnic and geographical origins technically lay within
this Persianate world: their ethnie is generally thought to have derived from
Iranian–Tajik populations31; and one could conceive of Ghur’s geography

as a “forgiving” zone permeably overlapping both the westernmost reaches

of the Indic sphere and the eastern peripheries of the Persianate. But
this borderland location combined with their own life-world along the
nomad–urban continuum, creating palpable socio-political, linguistic (cf.
above), cultural, and even epistemological distances between them and –
to adapt Ernest Gellner’s term – the “Great Tradition” of Perso-Islamic
elite lifeways.32 Indeed, as further discussed in Chapter 2, the topographic
and ecological realities of the region of Ghur apparently compounded
these conceptual (rather than physical) distances from the Persianate
ecumene, though in the end both the tangible and the intangible spaces
were surmountable.
As also described in detail in the Chapter 2, the Ghuris (and among
them the pre-imperial Shansabanis) were probably exposed to localized
practices of Islam – perhaps even observing some of its religious mores –
by the later eleventh century, if not before. It will be further argued that,
in tandem with their gradual but discernible rise to regional prominence
or “incipient” kingship, the early Shansabanis overtly adopted the trap-
pings of Islamic confessionalism – part of the “playbook” of kingship
prevalent in their specific milieu – in the first three or four decades of
the twelfth century. Perso-Islamic patterns of kingship were initiated and
sustained throughout the region via the precedents set by the Ghaznavids
and reinforced by the Saljuqs – between whose territorial ambitions fell
the home regions of the emerging Shansabanis (map, p. xv). Due to this
serendipitous location, the Shansabanis were cast in the role of inheritors
and eventual contributors to these rich cultural landscapes. In keeping
with the above-described, overlapping spheres of activity that permitted
the entry of outsiders into the complex of Perso-Islamic kingship, the
beginnings of the Shansabanis’ Islamic confessionalism were perceptibly
declared – and thus retrievable by later investigators – by means of their
patronage of architectural as well as other types of cultural production (cf.
esp. Chapters 2 and 3).
A century or so before the recordable emergence of the Shansabanis –
just after the establishment of the Ghaznavids (see below) – the Saljuqs
had entered Persianate frames of reference. Sources generally from outside
the Arabic–Persian scholarly traditions candidly assessed the eponymous
Saljuq ibn Duqaq’s embrace of Islam as part of a “design of conquest in the
Muslim world”: the historical texts’ prevarications regarding the reasons
behind Saljuq’s fateful choice might have served as historiographical syn-
ecdoche of the western Turkic populations’ uneven Islamic conversions
as they encountered Islamic socio-religious practices filtering eastward
along the byways of commerce, and met with again via the Saljuqs’ own
westward migrations.33 Although the specific details remain conjectural,
Islamic confessionalism and its manifestations became incumbent upon
the Saljuqs as they entered the Persianate realms. In ah 431/1040 ce, the

48 Iran to India
defeat of the Ghaznavid Sultan Mas‘ud I (r. 1030–1041 ce) at Dandanqan

by Saljuq ibn Duqaq’s grandsons Tughril (r. ah 431–455/1040–1063 ce)
and Chagri (ah 432–452/1040–1060 ce) meant that various constellations
of their descendants held sway across central–western Eurasia – initiating
prosperity as well as abundant building that would help define the archi-
tectural cultures of Iran and Central Asia until the Mongol incursions.
To their great advantage, the Saljuqs also inherited a sophisticated
politico-cultural landscape: it was replete with centuries of accreted
administrative infrastructure thanks to the multiple pre-Islamic imperial
formations that had spanned western–central Eurasia.34 Only the most
recent among these had been the Sasanians (third–seventh centuries ce),
whose “professionalization of a literate administrative class and the culture
of the court … were elaborations of what had gone before,”35 and in
turn were developed further according to the changing requirements of
ever-burgeoning elites during the eighth through eleventh centuries. To
be sure, the Saljuq sultans were notable patrons of architecture, but their
works were particularly aligned with proclaiming their Islamic confes-
sionalism (cf. below), with palace architecture serving very punctual func-
tions.36 In addition to the sultans, the upper echelons of military, political,
and scribal personnel throughout the empire – part of the Saljuqs’ inherited
administrative infrastructure – were also patrons to a significant degree.
The city of Isfahan had held geographical, strategic, and economic
significance at least since Sasanian times, becoming a centrally admin-
istered city – akin to a regional capital – for the Saljuqs in ah 443/1051
ce. It boasted a major masjid-i jami (congregational mosque) as of the
mid-ninth century, which would undergo expansion, renovation, and
even reconfiguration through the Safavid period.37 Given the city’s local
and transregional importance, Saljuq rulers and their courtiers continued
to patronize architecture and institutions throughout the city to maintain
its prosperity. Approaching our period of investigation, we should high-
light the interventions in Isfahan’s overall fabric during the late eleventh–
twelfth centuries, when major architectural projects attested to the vast
array of personages and impetuses underlying this patronage.
Sultan Malik Shah (r. ah 464–485/1072–1092 ce) directed his famous
vazir Nizam al-Mulk (ah 409–485/1018–1092 ce) to construct the Isfahan
jami’s south dome c. ah 479–480/1086–1087 ce (Figures 1.1 and 1.2), pos-
sibly in response to the new dome upon Damascus’s Umayyad mosque,
commissioned by the vazir Abu Nasir Ahmad ibn Fadl in ah 475/1082
ce after a fire. In turn, Nizam al-Mulk’s much younger political rival
Taj al-Mulk engaged in dangerous intrigue with the elder statesman to
win favor with the sultan, in part by commissioning a smaller but “in
many ways … aesthetically more successful” northern dome within the
already magnificent mosque (Figures 1.1 and 1.3). Not long after, the isfah-
salar Abi [sic] al-Fath ibn Muhammad commissioned a soaring minar

Figure 1.1 Congregational Mosque, Isfahan, Iran, founded tenth century CE, exterior:
view from east. © Photograph Alka Patel 2011.
Figure 1.2 Congregational Mosque, Isfahan, 1086–1087 CE, exterior: north dome.
(Figures 1.4–1.6), now known as Chihil Dukhtaran (“40 Daughters”), in

the city’s Jubara quarter (ah 501/1107–1108 ce).38
Other rungs of society emerged throughout Isfahan’s prosperous sur-
rounding areas. An “ascetic” (Pers. zahid) commissioned a sizeable minar
and contiguous mosque at the village of Gar, with the date of ah 515/1121
ce and his name inscribed in simple Kufic on the minar’s octagonal plinth.
And at the northern village of Sin, members of an important local family
of qadis patronized a mosque and minar (Figures 1.7–1.9) dated to ah
526/1132 ce and ah 529/1134–1135 ce, respectively – the minar bearing the
earliest epigraphically dated exterior use of blue-glazed tiles for the Kufic
bismillah at its current summit (Figure 1.9).39
Collectively, much of the enormous corpus of late eleventh–

50 Iran to India
Figure 1.3 Congregational Mosque, Isfahan, c. 1090 CE, interior: south dome
chamber. © Photograph Alka Patel 2011.
twelfth-century structures of the Iranian

Plateau constituted the cosmopolitan
expressions of what has been termed
“Saljuq-patronized” architectural
culture.40 Building upon “what had gone
before,” the array of public/“civic” and
palatial buildings subscribing to it were
individual and varied, sharing general
but uniquely expressed characteristics.
Nevertheless, the combination of salient
features circumscribing this architec-
tural culture is identifiable: baked brick
as the primary building material, with
an increasing use of blue-glazed tiles for
decoration; lofty hemispherical domes;
squat, pointed arches; emphatic portals
(pishtaq); extensive and stylistically varied,
usually painted, architectural epigraphy;
brick and stucco decorations of repeated
geometric and floral-arabesque iconogra-
phies (deriving from Abbasid-period
Figure 1.4 Minar known as Chihil stucco practices), also painted; and,
Dukhtaran, c. 1100 CE, Jubara quarter, Isfahan. where present, cylindrical and soaring,
© Photograph Alka Patel 2011. “pencil-thin” minars.41 For our purposes,

Figure 1.5 Chihil Dukhtaran minar, historical inscription panel. © Photograph

Alka Patel 2011.
this architectural culture expanded eastward with the Saljuqs’ imperial

ambitions.
The expansion of Saljuq imperialism from the principal cities of Rayy
and Isfahan predictably followed well-trodden routes of commerce, com-
munication, and pilgrimage (map, p. xv). Toward the east, the magnifi-
cently arresting elite accommodation known as Robat Sharaf (Figure 1.10)
attested not only to this eastward expansion, but to several other factors
besides: it was an important indication of the continued mobility of Saljuq
rule (cf. below) well into the twelfth century, and of the typological variety
encompassed within the prevalent architectural culture. Located on the
principal route connecting Marv and Nishapur, Robat Sharaf formed a
waystation for Saljuq officials and probably also the sultan and his imme-
diate entourage. Based on its epigraphic program, André Godard (1948)
proposed that the Robat was initially constructed c. ah 508/1114–1115
ce, perhaps even by Sharaf al-Din Abu Tahir ibn Sa‘d al-Din ibn ‘Ali
al-Qummi, who was governor of Marv and eventually Sultan Sanjar’s (r.
ah 511–552/1118–1157 ce) vazir. Other surviving epigraphic remains indi-
cated extensive renovations in ah 549/1154–1155 ce, possibly in the wake of
the destructive Oghüz rebellions against Saljuq authority, which resulted
in Sanjar’s capture and humiliating imprisonment between ah 548 and
551/1153 and 1156 ce.42 The Robat’s brick construction, courtyard–ivan con-
figuration (Figure 1.11), and the decorative repertory of extensive epigraphy
and floral–geometric motifs (Figures 1.12 and 1.13) altogether provide an
exemplar of non-religious building within the so-called Saljuq-patronized

52 Iran to India
Figure 1.6 Chihil Dukhtaran minar, topmost epigraphic bands. © Photograph

Alka Patel 2011.
architectural culture of Khurasan. Beyond such waystations, Saljuq polit-

ical power culminated at the major urban areas along principal routes of
travel, so that regional nodes such as Tus-Nishapur and Herat became
centers of their own radii of political and architectural–cultural dissemina-
tion (cf. Chapter 4).
As discussed in greater detail in Chapter 2, with Saljuq inroads into
the Balkh and Sistan regions as early as the mid-eleventh century,43 the
above-mentioned “Saljuq-patronized” architectural culture also found
manifestations in Afghanistan. The minar of Daulatabad, south of
Balkh (Figure 2.19), for example, subscribed to the cylindrical profile of
western Saljuq-period minars, but the Daulatabad proportions conveyed
a drum-like impression, rather than a “pencil-thin” one. Moreover, the
epigraphic program was monumentalized on the shaft, proclaiming its
patronage by the Saljuq vazir Abu Ja‘far Muhammad ibn ‘Ali (d. ah
515/1118 ce), the date of construction (ah 502/1108–1109 ce), and even the
stucco worker and carpenter.44 The general observations regarding pro-

portions and epigraphy are also applicable to

the known – though no longer surviving –
minar at Qasimabad, near Zahidan (before
ah 559/1164 ce) (Figure 2.18), about 240 km
southwest of Zaranj-Nad-i Ali (in modern
Iran’s Sistan and Baluchistan Province).
Qasimabad’s equally elaborate epigraphic
program apparently did not refer to Saljuq
authority, mentioning instead two of the
rulers of the local Nasri maliks of Nimruz/
Sistan,45 though the minar’s overall format
certainly derived from western architectural–
cultural ideas.
In the sense of coming from the
“outside,” both the Saljuqs and the emerg-
ing Shansabanis formed a contrast to the
hegemonic power of the Ghaznavids, already
entrenched in the Ghazna region by the end
of the tenth century. While the Shansabanis
eventually emerged from the anonymity of
central Afghanistan’s nomad–urban contin- Figure 1.7 Mihrab of mosque at Sin,
north of Isfahan, c. 1132 CE. © Photograph
uum, the Ghaznavids’ progenitors had been
Alka Patel 2011.
youths captured as slaves from Eurasia’s
eastern Turkic steppes. But thereafter, these
youths were effectively incorporated into
royal service farther west, absorbed into
and reared within the influential Samanids’
(ninth–tenth centuries) courtly ethos as
ghulams and important courtiers on the
rise (cf. Introduction). They were Turks
who had nonetheless become insiders of
the tenth-century Samanid court: the elite
environment par excellence, the crucible that
forged courtly styles and tastes especially via
Persian literary production (as well as Arabic),
casting long shadows over the following
centuries of cultural patronage and con-
sumption – the “Persianate millennium.”46
Indeed, we recall the supplanting of a purely
Turkic origin for Mahmud Ghaznavi’s
father Sabuktigin (d. ah 421/999 ce) with a
mixed Sasanian genealogy (cf. Introduction, Figure 1.8 Minar of mosque complex,
notes).47 As the Samanids waned and their Sin, c. 1134–1135 CE. © Photograph
ghulams increased their independence in the Alka Patel 2011.

54 Iran to India
Figure 1.9 Blue-glazed, topmost epigraphic bands of minar, Sin. © Photograph

Alka Patel 2011.
areas of Balkh, Bamiyan, and ultimately Ghazna, the new rulers largely
followed their erstwhile overlords’ royal protocols, administration, and
patterns of patronage.48
The Yamini-Ghaznavids’ kingly context included, however, ineluctable
specificities such as the material remains of previous political powers and
religio-cultural presences. Their co-optation of these pre-existing strata
is perhaps best demonstrated in the architectural remains surviving at
the royal sites of Ghazna and Bust-Lashkari Bazar. The grand palatial
architecture there (Figure 1.14) – discussed in detail in Chapter 5 – did not
diverge substantially from the non-religious buildings surviving through-
out the Persianate world, in fact echoing the courtyard–ivan plan at (for
example) Robat Sharaf, complete with intimate oratories tucked away
within royal ceremonial and reception spaces (Figure 1.12; compared with
Lashkari Bazar’s South Palace [Chapter 5]). However, monumental minars
(Figure 1.15) – towering above the central areas of the capital (Figure
1.14) and frequently encountered in other Afghan regions (cf. above and
Chapters 2–4) – were distinct at Ghazna. Rather than the unmistakable
“pencil-thin” cylinders marking urban fabrics in the central Saljuq lands,

Figure 1.10 Robat Sharaf, near Sarakhs, Razavi Khurasan Province, Iran, 1114–1115, 1154–1155 CE. © Photograph Alka Patel 2011.
14/09/21 5:38 PM
56 Iran to India
Figure 1.11 Robat Sharaf, second, inner courtyard. © Photograph Alka Patel 2011.
Figure 1.12 Robat Sharaf, oratory. © Photograph Alka Patel 2011.
the Ghazna minars were originally two-storied, with a bottom stellate shaft
supporting a tapering cylinder on top (cf. Figure I.20). Given the rarity
of stellate minars in the region – comparanda known only in Sistan (cf.
Chapter 2) – the Ghazna examples have incited debate as to their prece-
dents and/or inspirations.
The Yamini precinct of Ghazna lies within distant but visible view
of the composite Buddhist–Brahmanical Hindu site of Tepe Sardar (fl.

Figure 1.13 Robat Sharaf, intrados of arch with stucco decoration.

through c. ninth century ce),49 4 km to the southeast. Given the wide-

spread presence of Buddhism (and to some extent also the Shaivite branch
of Brahmanical Hinduism50) throughout Afghanistan into the ninth
century, W. Ball recently proposed the stellate plans to be “continuation[s]
of the Buddhist tradition.”51 Skilled workers crafted “stepped stellate”

58 Iran to India
Figure 1.14 Palace of Mas‘ud III, 1099–1115, Ghazni, excavation site, foundation
walls of buildings and apartments surrounding courtyard. © Photograph
Josephine Powell c. 1960. Harvard Fine Arts Library, Special Collections.
Figure 1.15 Minarets of Ghazni: minaret of Mas‘ud III, 1099–1115, and minaret
of Bahram Shah, 1118–1152, distant view. © Photograph Josephine Powell c. 1960.
votive stupas (Figure 1.16) for pious donors seeking spiritual merit at the
region’s already wealthy and powerful Buddhist monastic institutions, the
Tepe Sardar example (among the numerous other Buddhist monasteries
throughout Afghanistan) being one site located very near to the time and

Figure 1.16 The Buddhist site of Tapa Sardar, Ghazni, Afghanistan, third–ninth
century CE: upper terrace, star-shaped stupa (eighth century). F. Bonardi, neg.
7410–9. © Italian Archaeological Mission in Afghanistan, 1968.
place of Ghaznavid ascendancy. The rarity of Ghazna’s stellate minars, and

yet their proximity to such a prominent monastic establishment, renders
Buddhist precedents that much more plausible. The Yamini–Ghaznavids,
then, adapted to and even capitalized upon the cultural landscape they
inherited, to create distinct interpretations of the Persianate world’s cos-
mopolitan architectural cultures.
Notwithstanding the differences in geographical scale mentioned
before, a further resonance between the Saljuqs and the Shansabanis is
worth highlighting: the Saljuqs’ disconcertingly rapid and thorough per-
meation of the Persianate lands and Anatolia – and ultimately even the
very heart of caliphal authority in Baghdad by the mid-eleventh century –
forms an intriguing parallel to the profound and reverberating effects of the
Shansabanis’ eastward campaigns deep into the north Indian duab during
the last decade of the twelfth century. As noted in the Introduction and
argued in detail in Volume II, it was the Shansabanis’ politico-military and

60 Iran to India
architectural footprint in this region – more than the Ghaznavids’ plun-

dering raids – that irreversibly ushered the entire Indian subcontinent into
the Persianate world, and as one of the principal loci wherein Persianate
cultural production not only thrived, its practitioners also engaged with
Indic linguistic–cultural ambits for the following millennium.
These astounding successes – perhaps especially noteworthy for “out-
siders” like the Saljuqs and Shansabanis – all underscore mobility, the third
proposed sphere of activity of Perso-Islamic prestige cultures, which made
kingship accessible for ambitious newcomers to the Persianate ecumene. It
might appear evident that the Saljuqs’ and the Shansabanis’ origins within
nomadic and transhumant lifeways, respectively, and their consequent
seasonal movements predisposed them to regular traversal of large dis-
tances across contiguous yet diverse eco-cultural terrains. In turn, adapta-
tion to differing climates and topographies was a perennial feature of their
lifeworlds. Furthermore, flexible and consultative community and military
decision-making, and lack of investment in provisions and infrastructure
might have made nomad armies less reliable (see below), but, on the other
hand, the net positives were their virtually nil “operating expenses,” agility,
and quick recovery from defeats. Arguably, all of these lived experiences
coalesced into a form of “cultural capital,” enabling the expansive imperial
formations they managed to create across diverse landscapes.52
As noted above, the Yamini–Ghaznavids also conducted annual cam-
paigns for plunder deep into the north Indian duab throughout the first
half of the eleventh century, the later sultans reinitiating them for a time
during the century’s last decades.53 Although the Ghaznavids ostensibly
adhered to a sedentist and overall more traditional courtly model, their
India campaigns in particular could have nonetheless combined “mil-
itary purposes and seasonal migrations.”54 However, the Ghaznavids
never achieved a north Indian footprint akin to that of the Shansabanis,
whose “cultural capital” at least in part allowed them to succeed where the
Ghaznavids had not. In light of the historical trajectories and territorial
expansions of both the Saljuqs and the Shansabanis within the Persianate
world, it would seem that true mobility – without the rigidity and expense
of reigning courtly cultures – acquired a new significance for Perso-Islamic
kingship during the twelfth century.
Integrating Sources: A History of the Early Shansabanis
With a cognizance of the spheres of activity making it possible for

newcomers to the Persianate world to access kingship – specifically
Perso-Islamic kingship – we might more credulously envision the rapid
rise to elite status of a previously unknown group of Ghuris from central
Afghanistan’s nomad–urban continuum to become the Shansabanis of

the mid-twelfth century. As also noted above, however, the contemporary

authors examined historical vicissitudes while adhering to “historiograph-
ical expectations.” The unidirectional linearity of hindsight, for example,
is a characteristic to be borne in mind for all court chronicles (and no
less for modern scholarship), which at least partly explains the inclusions
and exclusions of events from textual narratives. The historical authors’
investments – to varying degrees – in highlighting their royal patrons in
the greater flow of history (such as it was conceived), and in “bending” the
constitutive events a posteriori into a concatenation, ultimately resulted
in the fabrication of a historical arc with a beginning, an apogee, and a
dénouement.
It comes as little surprise, then, that the known texts – including
Juzjani’s Tabaqat – leave many unknowns for the period of this volume’s
focus on the early Shansabanis in modern Afghanistan and Pakistan, that
is, prior to their entry into the north Indian duab. Partly in response to this
historiographical state of affairs, one of the fundamental premises of this
two-volume project is that architecture and material culture – especially
archaeological finds and numismatics – serve as indispensable primary
sources supplying evidentiary data, both enriching and at times apparently
contradicting their textual counterparts.
In the fifth and sixth decades of the twelfth century, the Shansabanis’
star seemed to rise steadily: their sphere of influence had rapidly extended
beyond their encampments and/or seasonal settlements in historical Ghur.
It is unclear precisely what prompted ‘Izz al-Din’s (c.1100–1146 ce) third
son Saif al-Din (c.1146–1149)55 to divide the ancestral territories – likely
definable by areas of transhumance and cultivation patterns (i.e., pastures,
agricultural areas, and dwellings56) – among his six brothers and himself
(Genealogy, p. xvii). But it would not be unfounded to suggest that a role
was played therein by customary laws governing property inheritance in
some nomadic societies, observable among modern groups.57
With the Ghaznavids and – especially after their defeat at Dandanqan
in ah 432/1040 ce – the Saljuqs both being distant powers whose author-
ity was manifest principally at important regional centers such as Zaranj
(Sistan),58 Herat, and Balkh, Ghur’s internal developments progressed
with only rare outside intervention.59 The overall outcome was a rapid,
multi-directional expansion of Shansabani dominion far beyond the
parameters of their seasonal transhumance (map, p. xvi), with a nominal
recognition of the larger powers’ suzerainty expressed mainly through
tribute and ceremonial acknowledgment, for example, in the khutba and
on coinage. Concurrently, the Shansabanis fractured into a loose confed-
eracy of independent lineages based at Firuzkuh, Bamiyan, and ultimately
Ghazna (see Chapters 3–5).
Juzjani’s consistent implication of the paramountcy of the Shansabanis
at Firuzkuh was surely overlaid on familial–political contestations. Some

62 Iran to India
hierarchy probably inhered in the designation of Qutb al-Din as malik

al-jibal – a grandiose title implying the designee’s importance.60 But at
the same time, Qutb al-Din was ‘Izz al-Din’s second son, not his first and
eldest, and neither he nor the eldest, Fakhr al-Din Mas‘ud, were born to
“the mother of the other Sultans.” Further, it is unclear whether Qutb
al-Din appropriated the title himself or, if not, why and when exactly it
was bestowed upon him. It is not impossible that the somewhat steadier,
commercially derived prosperity of Jam-Firuzkuh (see Chapter 2) enriched
and emboldened Qutb al-Din to adopt the title. Such an emphatic
self-elevation could justifiably have resulted from his otherwise secondary
status as the son of a mother from a lesser pedigree.61
The situation was similar with the Shansabani lineage of Bamiyan,
whose scions had come into possession of a region probably even more
prosperous than Firuzkuh, thanks to its centuries-long centrality within
commercial and pilgrimage networks (cf. Chapter 3). Several decades
before Juzjani recorded the Firuzkuh rulers’ concession of the title of
sultan to their Bamiyan cousins in ah 586/1190 ce, coins minted during
the reign of Fakhr al-Din Mas‘ud (ah 540–558/1145–63 ce) at the elusive
site of Parwan/Farwan, somewhere north of Kabul, claimed the title of
al-sultan al-‘azam for him.62 Furthermore, Juzjani reported that as the
eldest, Fakhr al-Din was not satisfied with his appanage of Bamiyan and,
toward the end of his reign, unsuccessfully attempted the usurpation of
Jam-Firuzkuh from his nephews Shams al-Din and Shihab al-Din – later
Sultan Ghiyath al-Din (d. ah 599/1203 ce) and Sultan Mu‘izz al-Din
(d. ah 602/1206 ce), respectively.63 From this admittedly fragmentary
evidence, it could nonetheless be ventured that Juzjani, in keeping with
the “historiographic expectations” of his time, attempted to salvage for
posterity a Firuzkuh-centered Shansabani dynasty from what was in reality
a periodic confederacy of disparate lineages. Often bitterly contentious in
their own time, the author elected instead to portray the Shansabanis in
a generally unitary light.64 Thus, the actual impetuses and contingencies
behind the reported events, along with their protagonists, are often to be
discerned in non-textual historical indices.
The unforeseen and successive deaths of the first three maliks of
Jam-Firuzkuh thrust to the forefront the dynamism of ‘Izz al-Din’s fifth
son, ‘Ala’ al-Din Husain (r. ah 544–556/1150–1161 ce). Motivated by
his brother Qutb al-Din’s death at the Ghazna court of Bahram Shah
(r. ah 512–547/1118–1152 ce) – even though a quarrel between Qutb al-Din
and his brothers drove him thence in the first place – Saif al-Din “brought
an army to Ghazna and captured it … and the first person of this clan to
take the title of Sultan was Sultan Suri.” Bahram Shah returned to Ghazna,
however, and executed Saif al-Din and his advisors in a most humiliating
way. In turn, Baha’ al-Din died of “excessive anxiety and grief for the death
of his brothers” while he was en route to Ghazna to exact vengeance.65

This last death provided ‘Ala’ al-Din an excuse he could not refuse.
Ostensibly a thirst for vengeance but also a defiant personality, and
doubtless a growing appetite for power and dominion – all underwrit-
ten by the Jam-Firuzkuh treasury replenished by the locality’s commercial
machinery – converged to impel ‘Ala’ al-Din Husain to deal a decisive blow
to the Ghaznavids. In ah 544/1150 ce, while his eldest half-brother Fakhr
al-Din Mas‘ud remained at Kashi (Bamiyan), ‘Ala’ al-Din set out against
the Ghaznavid sultan Bahram Shah and his forces to avenge the deaths of
his three older brothers, two directly and one indirectly at the hands of the
increasingly villainous Yaminis.66 According to Juzjani’s description of the
confrontation’s beginnings – unfolding upon the plains of Zamindawar,
where both forces had converged – Bahram Shah had calculated that the
mere mention of Indian elephants would deter ‘Ala’ al-Din from direct
military engagement.67 But the defiant Shansabani one-upped the Yamini’s
apotropaic threat, promising to match the elephants with the Kharmil,
which was in reality a human “weapon” in the form of two champion
Ghuri warriors whom ‘Ala’ al-Din ordered to bring down an elephant each.
After the Kharmils’ fulfillment of their ordered tasks and other dishearten-
ing setbacks, the Ghaznavid forces were defeated and Bahram Shah tempo-
rarily fled to India. Not satisfied with this defeat and still seeking thorough
vengeance, ‘Ala’ al-Din immediately marched to Ghazna. The destruction
he and his warriors unleashed on the Ghaznavid capital was also the fate of
the royal sites of Lashkari Bazar and Bust, which he plundered and burned
on his way back to Ghur from Zamindawar (cf. below).
Perhaps it was to be expected that, intoxicated with the avenging
victory over Ghazna and its rulers, ‘Ala’ al-Din Husain shortly afterward
openly challenged Saljuq authority, withholding the stipulated tribute
of the famous Ghuri war equipment (among other items). Although not
mentioned in Juzjani’s Tabaqat – but recorded in Ibn al-Athir’s Kamil –
during the course of protracted military engagements in the vicinity of
Herat, ‘Ala’ al-Din occupied this major commercial and cultural center
long enough to issue coins with the Herat mint stamp.68 But in a twist
of events worthy of dramatization, in ah 547/1152/3 ce the well-known
battle at Nab – a location between Chisht and Herat – determined ‘Ala’
al-Din’s foreseeable future: the large-scale defection of Oghüz, Khalaj, and
Turk horsemen from the Shansabanis to the Saljuqs brought about ‘Ala’
al-Din’s defeat and even imprisonment within Sanjar’s camp; although,
the Shansabani ruler’s quick wit earned him favor with Sanjar, who made
him one of his boon companions and eventually released him.69
The reign of ‘Ala’ al-Din Husain and its significance, particularly for
the Shansabanis of Firuzkuh, is gradually becoming clearer.70 Recently,
M. O’Neal (2016) presented coinage issued in the name of ‘Ala’ al-Din
Husain that was probably minted at Ghazna c. ah 556/1161 ce, perhaps
throwing a different light on this sultan’s intended ambitions for the

64 Iran to India
capital: the minting of coins there in his “second reign” – that is, the years
after his release from Saljuq captivity in ah 548/1153 ce until his death
in ah 556/1161 ce – strongly indicates that the sultan again besieged and
even occupied Ghazna toward the end of his rule. This may hint that
his earlier victory against Bahram Shah in ah 544/1150 ce had also been
intended as a conquest intended to integrate Ghazna within Shansabani
territories, rather than an act of avenging destruction. All in all, even
though ‘Ala’ al-Din’s violent revenge earned him long-term infamy and
the epithet Jahan-suz or “World Incendiary,” Juzjani likely described the
actual destruction at the sites with not a little hyperbole.71
Meanwhile, Bahram Shah eventually returned from India, and he and
his sons presided at Ghazna for another decade. But ‘Ala’ al-Din Husain’s
rampage had been both a symptom and an exacerbation of their decline:
the Ghaznavids’ hold over their own capital could not withstand the
violent Oghüz takeover in ah 557/1162 ce72 – after ‘Ala’ al-Din had again
briefly occupied the capital the year before (cf. above). The reign of the
last Ghaznavid sultan Khusrau Shah (r. ah 452–479/1160–1186 ce) passed
almost entirely in the Panjab. Thereafter, the Shansabanis effectively inher-
ited both the Ghaznavids’ opportunities and challenges, entering some
erstwhile Ghaznavid territories with little effort, while also provoking and
confronting the Saljuqs with mixed results (see Chapter 5).
‘Ala’ al-Din Husain’s wildly unpredictable reign – marked by successes,
failures, and dramatic recoveries – was of a piece with the unforeseen
events that elevated him to the Firuzkuh throne in the first place. Although
the untimely death of his older brother Sultan Baha’ al-Din had benefited
‘Ala’ al-Din, his position was unsafe due to the existence of claimants to
the Firuzkuh throne: Baha’ al-Din’s two young sons, Shams al-Din and
Shihab al-Din. Although ‘Ala’ al-Din had the boys imprisoned in the fort
of Wajiristan, fate restored their rightful prerogatives: upon ‘Ala’ al-Din’s
death, his son and successor Saif al-Din (r. ah 556–558/1161–1163 ce)
ordered the release of his cousins from their ten-year imprisonment, but
the latter’s own unexpected death in ah 558/1163 ce meant that the older
Shams al-Din was crowned Ghiyath al-Din, Sultan of Firuzkuh. Precisely
what ‘Ala’ al-Din had tried to prevent had nonetheless come to pass.
Sultan Ghiyath al-Din’s younger brother Shihab al-Din eventually
went on to great successes of his own, including the establishment of a
third Shansabani lineage (cf. map, p. xv; Genealogy, p. xvii; Chapters
4 and 5). Having captured Ghazna in ah 569/1173–1174 ce by means
of a joint effort with his older brother Sultan Ghiyath al-Din, Shihab
al-Din was eventually crowned Sultan Mu‘izz al-Din,73 and shortly
thereafter went on to consolidate the ex-Ghaznavid holdings throughout
Zamindawar. The younger Shansabani also “extracted Multan from the
hands of the Qaramatis [Shi‘a-affiliated Isma‘ilis]” in ah 571/1175–1176
ce. This was a momentous period for the Firuzkuh Shansabanis as well.

According to Juzjani, in the same year “the armies of Ghur and Ghazna
were prepared” for Herat: that is, possibly as reciprocation for support in
the Ghazna victory, Ghiyath al-Din called upon his younger brother to
join in the campaign. Their joint efforts were again successful, and the
Shansabani forces took Herat from Baha’ al-Tughril al-Sanjari, one of
Sultan Sanjar’s residual deputies. The occupation of this renowned city of
eastern Khurasan also led to the submission of the maliks of Sistan, “who
sent envoys and submitted themselves in unanimous service to the king
[padshah, i.e., Ghiyath al-Din].”74
At this juncture, it is worthwhile considering the consequences of these
victories upon the Shansabanis’ changing military manpower. Nomadic
contingents had been substantial within ‘Ala’ al-Din Husain’s forces when
he confronted Sanjar at Nab in ah 547/1152–1153 ce, and their defection
was at least partially responsible for the Shansabanis’ defeat. But with the
victory at Ghazna, the Shansabanis surely gained access to Khurasan’s
ghulam economy, fueled by the large-scale demand for Turkic youths
among all the ruling powers of the Persianate world; the old Yamini capital
had been one of the principal regional markets for this specialized military
labor.75 Although the city’s decade-long occupation by the Oghüz could
have reduced the supply of ghulams from the Eurasian steppe, circumstan-
tial evidence would indicate that the flow was not altogether interrupted:
after all, Mu‘izz al-Din’s eastward campaigns only grew in number and
distance after the Ghazna victory (cf. below), ultimately resulting in the
Shansabanis’ foothold in the north Indian duab thanks in large part to the
military skills and leadership of the well-known Mu‘izzi ghulams Qutb
al-Din Aibeg,76 Taj al-Din Yildiz, and Baha’ al-Din Tughril (cf. Volume
II).
The Herat victory – only about a year after that of Ghazna – would cer-
tainly have provided even greater access to the valuable military resource
of ghulams for westward campaigns, though these appear to have been less
successful than their eastward counterparts. Although Shansabani coinage
was minted at Marv (captured ah 597/1201 ce), and at Astarabad in Gurgan
(possibly ah 600/1204 ce), it is curious that no issues are known from
the pre-eminent Khurasani emporium of Nishapur, where a Shansabani
victory over the Khwarazm-shahi forces in ah 597/1201 ce led to the city’s
occupation until ah 601/1205 ce. Moreover, with the possible exception
of the tomb of the last Ghazna Shansabani ruler ‘Ala’ al-Din Muhammad
(d. ah 602/1206 ce) west of Nishapur at Bistam (Figure 1.17),77 there is no
known Shansabani architectural patronage beyond Herat.
In the east, the last of the Yamini–Ghaznavids had essentially been rele-
gated to Lahore as of the ah 570s/1170s ce, and Mu‘izz al-Din seemed to be
filling the politico-military vacuum left in their wake: in ah 574/1178 ce, he
undertook his first – albeit unsuccessful – long-distance campaign, prob-
ably through Kurraman (west of Peshawar) and again via Multan-Uchh,

66 Iran to India
Figure 1.17 Sheikh Bayazid complex, Bastam, Semnan Province, Iran: vaulted corridor ( Ilkhanid)
north of tomb of ‘Ala’ al-Din Muhammad of Bamiyan, d. at Ghazna 1206 CE. © Photograph
Alka Patel 2011.

to Anahilavada-pattana in northern Gujarat (map, p. xv).78 This was the

capital of the mighty Chaulukyas, a ruling house that was paramount
throughout northwestern India’s coastal and inland regions – mapping
onto the modern Indian states of Gujarat and southern Rajasthan –
between the later tenth through early fourteenth centuries. According to
Juzjani, the Chaulukya forces under either Mularaja II (r. 1176–1178) or
his young successor Bhimadeva II (r. 1178–1242) repelled Mu‘izz al-Din’s
attack, again in part with the use of war elephants.79
Evidently learning from this reverse, Mu‘izz al-Din and his forces sub-
sequently operated within shorter eastward radii. They conducted a suc-
cessful campaign to Peshawar in ah 575/1179–1180 ce. But, after a show of
force against the Ghaznavid sultan Khusrau Malik at Lahore in ah 577/1181
ce – the first of three such campaigns, as described below – Mu‘izz al-Din
again widened his circuit: in ah 578/1182 ce he marched to the Indus’s
débouchement at Dibul – now generally thought to be the archaeologi-
cal site of Banbhore80 – in southwestern Sindh, “captur[ing] the whole
of that seaside country and, taking much wealth, he ordered return [to
Ghazna].”81 It is possible that, in lieu of possessing a maritime port in
Gujarat, the Shansabani sultan aimed for neighboring Sindh: during the
twelfth century, the famed Dibul was still a major riverine emporium on
the Indus, serving as a crucial link between the Indian Ocean world and
the overland routes upriver – important enough, in fact, to be preceded by
at least one fortified bastion as the first line of defense from the Indus’s sea
access (see esp. Chapter 6).
Thereafter, it would appear that the status of the last Yamini–Ghaznavids
required a final resolution. For all future incursions into the north Indian
plains, unfettered possession of the important city of Lahore was neces-
sary, given its pivotal location in the center of the Panjab and in practi-
cally a straight line between Ghazna and the heart of the Ganga–Yamuna
duab.82 This meant the eradication of any lingering Ghaznavid loyalties
in the region. Thus, Mu‘izz al-Din’s second campaign to the Panjab in
ah 581/1185 ce left destruction in its path around Lahore, and the posting
of a loyal deputy at nearby Sialkot signaled the gradual cornering of the
Ghaznavids within their last stronghold.83
It is unsurprising that, in response to this obvious aggression, Khusrau
Malik mobilized his own forces to besiege Sialkot, forces that included
contingents of Kukhar tribesmen.84 Failure to capture the fort – and
thus no reward for their trouble – led the Kukhars to desert Khusrau
Malik. Largely defenseless, the last Ghaznavid ruler was finally captured
by Mu‘izz al-Din the following year (ah 582/1186 ce) and brought back
to Ghazna. As if yet another trophy of the momentous Shansabani
triumph over the Yamini–Ghaznavid house, Khusrau Malik was sent to
Sultan Ghiyath al-Din at Firuzkuh, where the minar commemorating
the capture of Ghazna already towered over the heart of the Firuzkuh

68 Iran to India
Shansabanis’ sard-sir royal encampment (cf. Chapters 3–5). Thence, the last
Ghaznavid sultan was dispatched to imprisonment and eventual death in
Gharjistan.85
Notes
1. Cf., e.g., Tapper 1997: 32–33. Fredrik Barth’s (1961) pioneering fieldwork
among the Basseri of southern Iran during the mid-twentieth century actually
revealed a polyglot milieu of Persian- and Turkish-speaking tribes, though
his data collection relied on oral reports of remembered pasts. These reports
were at times contradictory – e.g., regarding the tribe’s hoary origins (Barth
1961: 52) – but surprisingly consistent about recent history and the formation
of alliances and confederacies (Barth, 72ff., 86ff.). The Basseri’s tendencies
were to be contrasted to the Shahsevan, whose “different versions of origins
… reflected … differing constructions of their identity” (Tapper 1997: 317).
On the linguistic diversity among certain nomadic pastoralists in Iran and
Afghanistan, cf. Tapper 1983: 11–12. See also Khazeni (2009) on the Bakhtyari
of Iran, specifically the work known as Tarikh-i Bakhtyari (1910–1911), “a
sweeping combination … [of] oral histories and geographical lore”; and
Chapter 4.
2. See, e.g., Paul 2018b: 325–326 for the Oghüz nomads in Balkh and their
intersections with the Saljuq presence there, and the historical authors’ views
of “Khurasani history of this period in terms of a sedentary–nomad dichot-
omy, a perspective that continues to inform modern scholarly literature.”
Admittedly, the data of textual sources – additionally narrowed by their focus
on reigns, courts, and dynasties – can be supplemented with epigraphy and
numismatics. Nevertheless, mobile populations remain largely unrecoverable
even in this expanded catchment of information.
3. For the Saljuqs, see Peacock 2010: 53ff.; Peacock 2015: esp.13; Durand-Guédy
2013a, 2013b. Thomas (2018: 21–23ff.) also discussed the varying degrees of
erasure from the historical record of nomadic populations, often dismissed as
unruly and otherwise undesirable social elements – a theme running through
the works of the early Arabic geographers and into modern historical studies.
Cf. also Rao 1982 (mentioned infra); Thomas and Gascoigne 2016: 171ff.
4. Cf. Bosworth 1961.
5. See Baihaqi (trans. Bosworth) 2011, vol. I: 195, 198–207. Also Bosworth
1961: 126 and n. 22; Bosworth [1963] 2015: 118, 121, 227; Bosworth 1968a: 35;
Bosworth 1973: 121; Bosworth [1977] 1992: 68–69; Mahmud 2009: 59–60;
Inaba 2013: 78; O’Neal 2016a; Allegranzi 2019a, vol. I: 45; Rasikh 2019: ch. 2.
Juzjani, vol. I, 1963: 329 (trans. 1881, vol. I: 319–320) briefly mentioned raids
into Ghur also during the time of Sultan Mahmud’s father, the trusted
Samanid ghulam and military leader Sabuktigin (d. ah 389/999 ce), when the
latter had received the region from Tukharistan to Zamindawar as a reward
for his loyal service to his Samanid overlords. Cf. also Chapter 3 (notes).
6. Detailed in Deloche, vol. I 1980/1993: 29–32 and maps II and III; see also

Ruthven and Nanji 2004: 54–55. The position of Ahangaran in the northern
reaches of historical Ghur – see Chapter 2 – strongly indicates that some
Ghuris were intercepting caravans on the northern route from Herat toward
Balkh and beyond. Sultan Mahmud’s punitive campaigns there relatively
soon after his succession were surely motivated by commercial and political
factors. Ghaznavid control of Balkh and its surroundings was tenuous in the
early years, if we give credence to the Fa∂a’il al-Balkh, which states that it was
only as of the eleventh century that Ghaznavid-appointed qadis were posted
in Balkh. This more codified legal structure usually superseded the local,
public tribunal system that was customary in many parts of the Persianate
world; the process was, expectedly, often not free of conflict. Cf. Azad 2013:
118–119. Not only did the northern Herat–Balkh caravan route comprise an
important artery of mercantilism and communication, the zone was vital to
Mahmud’s control of Khurasan and his increasing independence from his
Samanid overlords (Bosworth 1973: 44–46). See also Paul 2018b: 319–320;
and de la Vassière 2018: 135–137 for the “gigantic [Kushan] military camp”
with a citadel north of Balkh that was likely most used in the Ghaznavid
period. The fortress of Jurwas, on the other hand, has not been conclusively
located, though its association with the sub-region of Darmashan could indi-
cate southern Ghur, toward Zamindawar (where Khwabin likely lay). Here
again, the above-mentioned caravan routes via Tiginabad/Old Qandahar
might have been just too tempting to resist for some Ghuris, particularly
during the over-wintering months of their seasonal migrations to the south.
Cf. Whitehouse 1976: 474ff. for the southern routes via Tiginabad/Old
Qandahar (see also Chapter 3). It should be remembered that the most direct
route of access to Ghur from Ghazna lay via Zamindawar – where Mahmud
left his three sons while on campaign in Ghur (Baihaqi [trans. Bosworth]
2011, vol. I: 195–196); clearly the Ghaznavid sultans felt compelled to secure
both the northern and southern routes from the Ghuris’ depredations.
7. If the French traveler J. P. Ferrier’s reports of the Turkman and Baluchi tribes
of Iran and Afghanistan in the mid-nineteenth century can serve as a remote
parallel for such activities in the historical past, it is worth considering that
the Ghuris also may have been after the people traveling with the caravans
rather than their goods. The desire for human commodities (according to
Ferrier, “man-stealing”) may be borne out by the flourishing slave trade
centered on historical Ghur, so often mentioned in pre-modern texts. See
Ferrier (trans. Jesse) 1976: 83–86; cf. also Khazeni 2012a: 141–143; and Næss
2015 (I am grateful to Dr. Tim Murray for the reference).
8. The phenomenon of mobile groups raiding caravans or even settlements and
cities is well known. There is an unfolding scholarly reconsideration of these
seemingly parasitic acts, however, as inevitably the impetuses for raiding dif-
fered according to specific groups and their circumstances. See further Næss
2015.
9. Cf. Baihaqi (trans. Bosworth) 2011, vol. I: 195, 198–207; and Bosworth 1968a:
35. It is noteworthy that mid-twentieth century Afghan–Pashtun nationalism
led the historian ‘Atiq Allah Pazhvak to propose a Pashto dialect as the histor-
ical language of Ghur (cf. Patel 2017: 151–152).

70 Iran to India
10. The image of a heavily fortified landscape and also Ghuris’ specialization in
special armaments (see below in main text) emerged already from Baihaqi’s
descriptions (trans. Bosworth), vol. I, 2011: 199–204, vol. III: 97–98 (note 446
to ah 421/1030 ce).
11. Cf. Bosworth (1961: 118, 124; Bosworth [1977] 1992: 68), who cited Ya‘qut,
and relied largely on Marquart’s observations, in turn gleaned from al-Biruni.
12. Mahmud 2009: esp. 242ff.
13. Mahmud 2009: esp. 251–254; also Hunter 2010: 82. See Patel 2017 for the
historiographical context and analysis of Mahmud’s monograph. The histor-
ical texts’ frequent mention of the Ghuris’ arms manufacture further under-
scores their undoubtedly symbiotic relationship with settled society, echoing
Gellner’s observations above. See also Thomas and Gascoigne 2016: 174ff.;
Thomas 2018.
14. See O’Neal 2015; O’Neal 2016a. Ghafur (1960: 1–12) also provided an over-
view of the contemporary and later textual sources with information on the
Shansabanis, though adopting Barthold’s (1968: 30) characterizations of
“the second half of the 12th and the beginning of the 13th century [as] one of
the darkest pages of Muslim history,” and the text-based data as contradictory
and requiring “very critical study” (borne out by Bosworth 1977/1992: 115ff.).
Ghafur’s useful survey of the contemporary works, including those unfor-
tunately lost and known only through later authors’ citations, encompassed
both the chronicles patronized by other courts and containing some mention
of the Shansabanis (e.g., al-Rawandi’s Rahat al-Sudur wa Ayat al-Surur,
c. 1205, principally focused on the Saljuqs), and those written within the
Shansabani courts themselves, such as Fakhr al-Din Razi’s Risala-yi Bahaiyya
(see also Chapter 3), dedicated to the Bamiyan ruler Baha’ al-Din Sam (r.
1192–1206) (cf. infra in main text). However, given the evident participation
of the non-extant works and their authors in the Persianate literary world
and its conventions of poetry and prose – an assumption bolstered by the
surviving texts – it is practically certain that the overall corpus offered little
more data, perhaps only allusions to the Shansabanis’ pre-imperial nomadic
ancestors and lifeways.
15. See Asif 2016: 51–55, 57–60; Anooshahr 2018b: 2–3.
16. I borrow the phrasing from Anooshahr (2018b: 2 and n. 3), whose work
focused on the fifteenth- and sixteenth-century emergence of the great
Eurasian empires of the Ottomans, Safavids, and Mughals. Anooshahr’s
incisive analyses of the crafting of their identities also informs parallel pro-
cesses in the tenth–twelfth-century period of nomadic confederacies coming
to rule large swaths of the Islamic world, most notably the Saljuqs. See also,
inter alia, Bosworth 1968a: 40–41; Bosworth 1973: 61–62; Meisami 1993; and
Chapter 3.
17. See Bosworth 1973: 55 and passim; esp. Meisami 1999: 20–21, 37–45; also
Peacock 2018: 4ff.
18. Cf. Bosworth 1961: 125–126 and n. 32. The transformation of Zahhak from the
tyrant and sinner of canonical texts – such as al-Ghazali’s (d. ah 504/1111 ce)
Nasihat al-muluk (cf. Lambton 1981/2006: 122–123) – to a desirable ances-
tral figure perhaps best demonstrated the great epistemological/conceptual

distances between mainstream Persianate trends and the largely isolated pop-
ulations inhabiting the hinterlands of even renowned urban centers such as
Ghazna – further exacerbated in the case of Ghur: Juzjani’s erudition must
have had to concede to his patrons’ decidedly provincial perceptions.
19. See Juzjani, vol. I, 1963: 320 (trans. vol. I, 1881: 301ff.). Cf. also Meisami’s
(1993) study of the process by which pre-Islamic Iranian mythology and
specifically Islamic sources together provided origins during the eleventh
century.
20. See esp. Volume II for the texts patronized at the north Indian courts, which
have little to offer on the early Shansabanis in Afghanistan. Kumar (2007:
21–22, 92, 297) addressed this issue, though more specifically regarding north-
ern India’s Persian literary culture during the early thirteenth century, which
tended to be “Delhi-centered” and “wedded to the history of a unitary state
formation.” Even Uchh Sharif, the contemporaneous locus of political power
and cultural capital, and the site of Nasir al-Din Qabacha’s court in upper
Sindh, patronized its own cohort of scholars and poets and witnessed the
circulation of these literati – most notably Juzjani himself – between the two
rival cities. See Alam 2003: 138–141.
21. Due to the confusion introduced by E. Denison Ross (1922), it bears reitera-
tion that this personage must be distinguished from Fakhr al-Din Muhammad
ibn Mansur Mubarak-shah al-Quraishi (‘urf Fakhr-i Mudabbir), a Persian
litterateur who began his career at Ghazna under the later Ghaznavids. The
latter’s decline and eastward retreat prompted the writer’s own emigration to
Lahore and eventually Delhi. No doubt in part due to overlapping profes-
sions, careers, and lifespans (and, no doubt, names), Fakhr-i Mudabbir was
initially confused with the genealogy’s author Fakhr al-Din Mubarak-shah.
Cf. Juzjani vol. I, 1963: 318–319 (trans. vol. I, 1881: 301–302). Also Khan 1977;
Siddiqui 2010: 17–18; Bosworth 2012a; Auer 2012: 22–23; Auer 2018: 96 and
n. 5.
22. Juzjani, vol. I, 1963: 318–320ff. (trans. vol. I, 1881: 300–302ff.)
23. See esp. O’Neal 2016a; O’Neal 2015; O’Neal 2020; Cribb 2020. I am grateful
to both authors for sharing advance copies of their forthcoming publications.
24. Cf. Rasikh 2019: ch. 3. The various genres of Persian prose and poetry in
India largely derived from the broader Persianate world of the time (Kumar
2007: 366; Alam 2003). The political and cultural specificities of the Indian
location nonetheless filtered into even the universal history or tarikh tradi-
tion, to which Juzjani’s Tabaqat belonged. Despite Tabaqat’s multifaceted
representation of events (cf. Siddiqui 2010: 93ff.), Kumar (2007: 367) dis-
tinguishes the work as an atypical tarikh, highlighting its grouping of people
according to “social affinity” – in itself perhaps a subtle but firm erasure of the
Shansabanis’ “lowly” non-sedentist origins.
25. Cf. Hodgson (esp. vol. II, 1977: 293–94) for the coining and explication of the
term. For a recent engagement with the concept, see esp. Green (2019: esp. 7),
who undertook its transtemporal and spatial interrogation, proclaiming “the
need to analytically denaturalize Persian’s [as a language] civilizational ties to
Islam and denationalize its primordialist ties to Iran.”
26. As symptomatic of the current scholarly status quo, Spooner (2019: 305)

72 Iran to India
observed that “nomadic tribal communities have been an important factor in

Islamic and Persianate history down to the present,” but went on to retain the
distinction between “Arab (tribal, Sunni) and Persian (urban, Shi‘i)” (ibid.,
308). Rather, re-examining these broad and largely artificial divisions requires
investigation of how nomadic groups in fact transcended them – a transcend-
ence amply evidenced throughout the “the Persianate millennium.”
27. Cf. Biran 2004; Biran 2005: 94ff., 196 and passim; Bosworth 2010; McClary
2020: 7ff.
28. Bosworth, 1961, 1965, 1973, 1992; Pazhvak 1968: 90, 94–95; O’Neal 2016a.
29. Thomas (2018: 29) rightly eschewed the otherwise perspicacious C. E.
Bosworth’s historically anachronistic term of “buffer state” to refer to the
coalescing Shansabanis of the first half of the twelfth century: the phrase was
first used only in the late nineteenth century to describe the status of the then
Durrani–Pashtun kingdom of Afghanistan vis-à-vis Britain and Czarist Russia
in the so-called Great Game; moreover, while the Shansabanis’ homelands of
seasonal transhumance were technically sandwiched in between the Saljuqs
and the Ghaznavids, their early territorial presence was far from bounded and
fixed, as the modern term would imply. Nevertheless, the term was used again
to refer to the Balkh region in the seventh century: cf. de al Vassière 2020: 127.
30. Cf. Durand-Guédy 2013a, 2013b. Following the same author (2018), it is
important to highlight the integration of the luxurious trellis tent (Pers.
khargah) – initially associated with “Turanian” or Turkic contexts of the
Eurasian steppe lands – as a part of the visual and textual description of power
in western Eurasia’s Perso-Islamic courts by the eleventh century. While this
incorporation surely signaled mobility and agility as a part of royalty (cf. infra
in main text and Chapter 3), “[i]n Iran the khargah remained exclusively a
status symbol … not imply[ing] a change of lifestyle: the urban location of
the Iranian court” (ibid., 835). See also infra in main text and notes.
31. Bosworth 1961: 118.
32. See Gellner 1983: 446; also Introduction and Chapter 4.
33. Peacock 2010: 122–23; Peacock 2015: 25ff. See also this author’s discussion
of the Christian author Bar Hebraeus – who drew upon the Malik-nama
tradition, like his counterparts of the Perso-Arabic histories – and therein
the motivations behind Saljuq conversion to Islam (ibid., 246–247). A par-
allel Muslim-convert progenitor figure was Satuq Bughra Khan (mid-tenth
century) of the Qarakhanids (cf. Bosworth 2010: 22).
34. Cf. Peacock 2015: e.g., 39, 48.
35. Spooner 2019: 301.
36. See Durand-Guédy 2013b for the Saljuq period’s “clear break with the
past” especially in terms of the spatial organization of power. Cf. also
Chapter 3.
37. See esp. Blair 1992: 160–161; and Lambton and Sourdel-Thomine 2008: 167–
170, 172–174.
38. For the Isfahan jami‘, see Blair 1992: 160–167; Hillenbrand 2015. For the
example of the Chihil Dukhtaran minar, cf. Smith 1936: 318–323; Y. Godard
1936: 361–363.
39. For Gar, see Smith 1936: 323–337; Y. Godard 1936: 363. For Sin: Smith 1936:

327–331; Smith 1939; Y. Godard 1936: 363–364; Miles 1939: 11–14; Bloom 1989:
160.
40. Cf. Introduction. Despite Sourdel-Thomine’s own reservations, she gener-
ally leaned toward such dynastic appellations, both for the Saljuqs and the
“Ghurids” (cf. Sourdel-Thomine 1953; Sourdel-Thomine 1960: esp. 277).
However, the author did clarify early on that, for the Saljuqs “the term did
not suppose a strictly dynastic remit, but designated … all the constructions
erected at the time when Saljuq power played a primary role in the Islamic
lands” (1953: 109 n. 3) – a useful clarification since, as we see in the main text,
Saljuq-affiliated elites and other personnel were robust and visible patrons.
See also Hillenbrand 1994: 152. The architectural culture under discussion has
also been described as the “wider architectural koine” specifically in reference
to Qarakhanid-patronized buildings (McClary 2020: 108).
41. Peacock 2015: 248.
42. See Godard 1949: 10–11, 31, 54 and passim; and Kalus, TEI: Nos. 34503, 34495,
34499, 34491, 34493, 34497, 37426, 40834, 34501; Peacock 2015: 107–109.
43. Cf. Bosworth 1994: 381–386; Peacock 2015: 35–36, 41–42; Paul 2018b; see also
infra in this chapter.
44. Sourdel-Thomine 1953: esp. 122–129.
45. See Tate 1910/12: facing 22, 26–27; esp. O’Kane 1984: 89–97; Bosworth 1994:
395–396.
46. Meisami 1999: 20ff.; Green 2019: 12–13, 16; Spooner 2019: 310ff.
47. By contrast, the Saljuqs “vaunted their [Turk] origins through their alien
Turkish names, and introduced new political symbols and practices that orig-
inated on the steppe” (Peacock 2015: 3). This difference might be explicated at
least in part through the differing historical origins of the two ruling groups:
The Ghaznavid progenitors’ and early sultans’ memories and knowledge of
their natal world, the Eurasian steppe, would be at best faint, given their
steeping in the Perso-Islamic prestige culture of the Samanid court; by con-
trast, the un-enslaved and free Saljuqs’ active infiltration into the Persianate
world allowed less compromised memory and cultural pride to travel west-
ward with them.
48. See esp. Bosworth 1968a: 36ff.; Bosworth 1973: 61; Peacock 2015: 3. See also
Chapter 3 (notes) for Sabuktigin’s enduring loyalty to the Samanids.
49. For fascinating “hybrid” iconographies combining Buddhist and Brahmanical
elements at Tapa Sardar near Ghazna, see Taddei 1968; Taddei and Verardi
1978; Verardi and Paparatti 2005. The proximity of the Buddhist–Brahmanical
site to the east of the great capital of Ghazna led scholars to consider the
potential interaction between the later residents of Ghazna and the aban-
doned pre-Islamic ruins: Tapa Sardar could have been the Shahbahar or
Shahrabad of Ghaznavid sources, mentioned by Baihaqi and Gardizi, and in
the poetry of Farrukhi and Ansari: located a farsakh from Ghazna in the plain,
troops were reviewed here before their departure on ghazwa to India. Cf. also
Bosworth [1977] 1992: 56. Linguistically, Shahbahar or “the king’s temple”
[?] makes reference to the Buddhist vihara, often incorporated into Arabic
toponymic suffixes as bahar. Cf. Taddei 1968: 109–110; Melikian-Chirvani
1974: 8–10; Allegranzi 2014: 109–110; Errington 2017: 44. According to a more

74 Iran to India
recently proposed chronology, Tapa Sardar was probably abandoned in the

eighth century; rather than a sudden destructive episode such as an Arab raid
causing its abandonment, Verardi and Paparatti (2005: 441–442) described
a more gradual process set off by the Chinese Tang rulers’ divestment from
the region and the consequently more impactful Arab campaigns there. Cf.
also Rezakhani 2017: 183–184. The lapse of only two centuries between Tapa
Sardar’s disuse and the establishment of the Ghaznavid dynasty offers intrigu-
ing possibilities regarding the awareness and perhaps reuse of the Buddhist–
Brahmanical site by the residents of nearby Ghazna.
50. As of at least the seventh century ce, various sects of Brahmanical Hinduism
attracted growing numbers of adherents in eastern Afghanistan, becoming
more evident in archaeological finds from Kabul and its vicinity, attributable
at least in part due to the westward extension of the Sahis of Kashmir and
Panjab. Cf. Rehman 1979: 33–34; Klimburg-Salter 1989: 55–56; Klimburg-
Salter 2008: 132; Baker and Allchin 1991: 10ff.; Errington 2017: 43. Studies
of Jaina metal sculptures found in the vicinity of Kabul and dating to the
sixth and twelfth centuries have provided evidence of Jainism in eastern
Afghanistan, likely in the form of traveling Jain merchants: stylistically the
recovered sculptures seem to originate in northwestern India (Pal 2007/2012).
It is possible that the singular find of a broken marble jina in the Bamiyan
valley may have been locally carved, given the marble’s similarity to that of
the Ghazna tombs (see Volume II), though the stone also calls to mind the
great marble deposits in the Gujarat–Rajasthan region. Cf. Fischer 1962.
51. Cf. esp. Ball 2020 (forthcoming); for Tepe Sardar, see Ball 2019: No. 1180 for
the most up-to-date bibliography.
52. Cf. Irons 2003; Thomas 2018: 26; Malagaris 2020: 36–38, 51–52. See also infra
in main text and Chapters 3, 5 and 6.
53. Cf. Bosworth [1963] 2015: 75–76 and passim; 235; Bosworth [1977] 1992:
61–67; Pinder-Wilson 2001: 165–166. See also Chapter 4.
54. Inaba 2013: 89ff.
55. The genealogy of the Shansabanis recently reconstructed by O’Neal (2016a)
makes it possible to associate dates with reigns beginning at the dawn of the
twelfth century ce (ah end fifth–sixth centuries). At least some of the reigns
and sequences of events can be verified from parallel sources, most notably
the ascendant Shansabanis’ interactions with the Saljuqs, and the waning arc
of the Ghaznavids in Zamindawar: in addition to Juzjani, twelfth- through
fourteenth-century sources include Nizami ‘Aruzi (fl. early twelfth century),
Ibn al-Athir (1160–1233), Juvaini (1226–1283), and Rashid al-Din (1247–1318).
Cf. Ghafur 1960.
56. This extrapolation is derived from studies of the imperial Saljuqs who, even
after Tughril’s adoption of the title Sultan (1038) and Malik Shah’s (r. 1053–
1092) move from Rayy to Isfahan as his seat of political power, “the dynasty
pursued an itinerant lifestyle, not from town to town, but rather from pasture
to pasture” (Durand-Guédy 2013a: 172).
57. Cf. Juzjani vol. I, 1963: 335–336 (trans. vol. I, 1881: 339–340) for Saif al-Din
Suri’s division and distribution of patrimony among his brothers. Barfield
(1993: 100–102) observed that modern nomadic pastoralists in southwestern

Asia based succession on the agnatic line, and tended to regard the single
household or tent as the basic socio-economic unit, rather than a “joint-fam-
ily ideal” where males “pool their labor and animals under the direction of
their father.” Thus, property would be fractionally divided upon a patriarch’s
death, effectively constituting a system of “anticipatory inheritance” (Barth
1961: 19–20). It has additionally been observed that, while these groups might
claim to adhere to Qur’anic stipulations of property division, the realities of
agnatic inheritance generally excluded females; only in cases of agnate con-
flict would religious authorities be invoked, who tended to be more exacting
about Qur’anic adherences while adjudicating disputes (cf. Barfield 1993:
101). Another prominent historical example of nomadic anticipatory inher-
itance was the division of existing and future conquests between Saljuq ibn
Duqaq’s grandsons Tughril and Chaghri in the early eleventh century ce
(Peacock 2016: 6).
58. Although the exact political hierarchy of the Sistan Maliks vis-à-vis the
Shansabanis is not clear, Bosworth’s view of Sistan overall was that the region
was caught up in the larger politico-military developments in Khurasan, being
subject to the Ghaznavids and the Saljuqs throughout the eleventh and mid-
twelfth centuries, eventually ending up as Shansabani vassals upon accession
of Sultan Ghiyath al-Din to the throne of Firuzkuh (ah 558/1163 ce). They,
like the Shansabanis, succumbed to the Khwarazm-shahs by c. ah 612/1215 ce.
See Juzjani vol. I, 1963: 395–396 (trans. vol. I, 1881: 447–448); Bosworth 1994:
398–399; O’Neal 2013: 58–59.
59. The one documented example of this external interference resulted from
Mahmud Ghaznavi’s punitive campaign to Ahangaran in Ghur in ah
401/1011, to discipline the Ghuris’ caravan raiding: according to Juzjani (vol.
I, 1963: 329–330 (trans. vol. I, 1881: 320–329), the sultan removed Muhammad
ibn Suri and his older son Shish in favor of the more pliant and younger son
Abu ‘Ali.
60. O’Neal 2016a.
61. Cf. Juzjani vol. I, 1963: 335, 385 (trans. vol. I, 1881: 338–339; ibid., 422);
Edwards 2015: 94. Saif al-Din Suri’s designation as distributor of patrimony
was itself indicative of the cultural preference for sons of lawful wives –
presumably women of commensurate ancestry – as technically he was only
‘Izz al-Din’s third son. Among the seven male children (no mention being
made of females in this generation), the eldest Fakhr al-Din was born to “a
Turki servant,” while the second Qutb al-Din also to “a woman … of no high
name … the doorkeeper and servant of the mother of the other sultans …”
Such hints seem to indicate that a melding of the Shahsabanis’ (and likely
other Ghuris’) customary laws of inheritance with the more orthodox Islamic
laws pertaining to dispensation of patrimony had not yet taken place. See
also Chapter 3. The Shansabanis may never have come to follow strict Islamic
orthodoxy in patrimonial inheritance: upon Ghiyath al-Din’s death at Herat
in ah 599/1203 ce, Mu‘izz al-Din, now the eldest male, also proceeded to
distribute appanages to his own male agnates, as his uncle Saif al-Din had
done more than a half-century earlier. Cf. Juzjani, 1963 vol. I: 335, 385 (trans.
vol. I, 1881: 471–472). Thus, even after the Shansabanis’ public espousal and

76 Iran to India
patronage of Sunni orthodoxy (the Shafi‘i madhhab at Firuzkuh and Herat,

the Hanifi at Ghazna; cf. Chapter 4) age-old customary inheritance laws
apparently remained in place.
62. Cf. Cribb 2020: 122, esp. for a discussion of the location of Parwan/Farwan;
and O’Neal 2020: 202–204, 209–210.
63. Cf. Juzjani vol. I, 1963: 335 (trans. vol. I, 1881: 371–372 and passim, 424–425);
also O’Neal 2020: 201, 207.
64. Cf. also O’Neal 2020: esp. 210.
65. Juzjani vol. I, 1963: 338 (trans. vol. I, 1881: 343).
66. Cf. Juzjani vol. I, 1963: 337, 338, 393–395 (trans. vol. I, 1881: 340–343,
439–445); Bosworth [1977] 1992: 113–115. See also infra in main text; and
Chapter 4.
67. Cf. Juzjani vol. I, 1963: 341–342 (trans. vol. I, 1881: 350–351). Bahram Shah’s
expectation that his threat of elephants would suddenly make ‘Ala’ al-Din
capitulate had appeared cryptic until recently: the elephant’s diabolical asso-
ciations within Islam had to be reconciled with their increasing numbers in
the Ghaznavid arsenal, where they became invaluable destructive and tacti-
cal weapons. Mahmud had attempted to distract from his reliance on ele-
phants by very propagandistically raiding and destroying India’s “infidel” idol
temples during the latter part of his reign. Cf. Anooshahr 2018b. However,
it is possible that a century later, Bahram Shah still thought the threat of
elephants would frighten his less worldly foe (prior to the Shansabanis’ own
campaigns to India; cf. Volume II), or perhaps he assumed the elephant’s
diabolic associations lingered among the Shansabanis in light of their more
recent and thus lesser Islamization, accompanied by an incomplete literacy in
Islamicate culture.
68. According to additional numismatic evidence, ‘Ala’ al-Din’s son and succes-
sor Saif al-Din (r. ah 556–558/1161–1163 ce) also successfully occupied the city
in ah 557/1162 ce, at least for a short period before his death (see Chapter 4).
Cf. O’Neal 2016a; O’Neal 2020: 199–200.
69. Cf. Juzjani vol. I, 1963: 346–348 (trans. vol. I, 1881: 357–361); Leshnik 1968:
39; Bosworth [1977] 1992: 118–119; O’Neal 2016a. For the Saljuqs’ nomadic
military power, see Peacock 2015: 50, 143, 168, 224. See also infra in main text.
70. Another notable “conquest” during the reign of ‘Ala’ al-Din Husain was
that of Gharjistan, which was apparently effected through a marital alliance:
‘Ala’ al-Din married Hurr Malika, the daughter of the reigning shar (not
unlike the hereditary title of the Shah of Bamiyan; cf. Chapters 3 and 4),
who was nonetheless only “one of the Maliks of Gharjistan.” See Juzjani
vol. I, 1963: 349 (trans. vol. I, 1881: 363). It was only during the reign of his
famous nephew Shams al-Din ibn Sam (eventually Sultan Ghiyath al-Din,
r. 1163–1203), however, that the magnificent madrasa of Shah-i Mashhad was
constructed in Gharjistan (Chapter 4).
71. See Juzjani vol. I, 1963: 343–344 (trans. vol. I, 1881: 353–356); Bosworth [1977]
1992: 117–118; also O’Neal 2016a. Despite Juzjani’s description of ‘Ala’ al-Din
Husain’s destruction of Ghazna and Bust-Lashkari Bazar as reaching legend-
ary proportions, it has been archaeologically ascertained that these sites were
still salvageable in the 1170s, even after the decade-long Oghüz encampments

at Ghazna in the 1160s. A little more than twenty years later the newly
crowned Sultan Mu‘izz al-Din went on to add his own mark to Ghazna and
Lashkari Bazar, with architectural patronage consisting of palace renovations
and other works. Cf. Chapter 5.
72. Cf. Bosworth [1977] 1992: 124–125; O’Neal 2016a.
73. O’Neal (2016a) has pointed out that the precise moment of the younger
Shansabani’s adoption of the regnal laqab is not clear: not only is there
numismatic evidence for continued usage of “Shihab al-Din,” but this laqab
was frequently used to refer to the younger Shansabani in north Indian
Sanskrit inscriptions (cf. Prasad 1990: 4, 8, 12, 28, 30). By contrast, “Mu‘izz
al-Din” was used in the Perso-Arabic inscriptions on the Qutb mosque – esp.
the northern entrance – and on the first story of the Qutb minar (Horovitz
1911–1912: 14–18). See Volume II.
74. Cf. Juzjani vol. I, 1963: 357–358, 396 (trans. vol. I, 1881: 377–379, 449).
75. Cf. Barthold 1968: 237–238, 329–330, 338–339; Bosworth [1963] 2015: 98–102;
Kumar 2007: 77–79; Peacock 2015: 217ff.
76. As described by Fakhr-i Mudabbir (in Ross 1922: 397ff.), a qadi named Fakhr
al-Din Kufi purchased Aibeg at the market of Nishapur, where the flow of
Turk slaves was even more active than in Ghazna’s market. Aibeg thereafter
passed into the Mu‘izz al-Din’s possession. It is all the more surprising, then,
that there is no known Shansabani coinage minted at Nishapur (cf. infra in
main text).
77. Cf. O’Neal 2016a; and O’Neal 2015; O’Neal 2020. Despite the last Shansabani
ruler’s burial at the Sheikh Bayazid complex at Bistam, at present it is unas-
certainable whether the tomb structure was actually constructed by or for
him. Cf. also Adle 2015: 242–243; and Volume II.
78. See Juzjani vol. I, 1963: 397 (trans. vol. I, 1881: 451). Juzjani’s mention of
Multan-Uchh as midway points from Ghazna would point to eastward passage
into the Indus regions via the Gomal Pass – cf. Deloche (trans. Walker) vol.
I, 1993: 26–27. However, for at least a century since the Ghaznavids’ eastward
campaigns, and well into the thirteenth century, the favored route between
Ghazna and the middle through upper Indus areas appears to have been via
Kurraman – a major mint site for the Ghazna Shansabanis’ jitals. Cf. Tye and
Tye 1995: Nos. 138, 174–178; and esp. O’Neal 2015.
79. While Juzjani (vol. I, 1963: 397 (trans. vol. I, 1881: 451)) named Bhimadeva
– i.e., Bhimadeva II, r. 1178–1241/2 – as the triumphant foe, the reign of his
predecessor Mularaja II (r. 1176–1178) also overlapped with the date of the
Shansabani campaign. In the corpus of Chaulukya copper-plate inscriptions,
the imperial genealogies (Skt. vamsavali) of at least three major land grants
credited Mularaja with the defeat of the garjjanakas or mlecchas (among the
many labels for the northwestern invaders in Indian inscriptions and other
sources, as summarized for the eighth through early seventeenth centuries by
Chattopadhyaya 1998: 92–97). See Patel 2004: 5–6 and notes; and Volume
II. While it is quite possible, then, that Mularaja II in fact led the Chaulukya
forces repelling this first Shansabani campaign far beyond the Indus, Juzjani’s
substitution of Mularaja’s successor Bhimadeva (II) in his recounting of the
episode was not only an easy confusion – one’s reign ended and the other’s

78 Iran to India
began apparently in the very year of Mu‘izz al-Din’s incursion – it also

heightened a historical parallel: Mahmud’s oft-cited Somnath campaign
of ah 416/1025 ce confronted the Ghaznavid sultan with the Chaulukya
maharajadhiraja Bhimadeva I (r. c. 1022–1064 ce). Cf. Bühler 1877: 186,
213; Trivedi 1994: 165. As further discussed in Volume II, during this defen-
sive campaign against the Shansabanis the Chaulukya forces were militarily
aided by their Nadol Chahamana feudatories. Cf. Kielhorn 1907–1908: 72;
Bhandarkar 1911–1912: 71–72.
80. Although the archaeologists who initially excavated at Banbhore remained
circumspect regarding its identification as Dibul (e.g., F. Khan 1963: 50), the
site is now viewed to be that of the pre-eminent Indus port center (A. Khan
2003: 2). Cf. also O’Neal 2013: 64–65; and infra in this chapter.
81. Juzjani vol. I, 1963: 397 (trans. vol. I, 1881: 452).
82. Cf. esp. Deloche (trans. Walker) vol. I, 1993: 30–34.
83. The strategy of obtaining command of a fort near a targeted stronghold, and
placing a loyal deputy there awaiting further orders – in this case, Husain
ibn Kharmil – very much echoed the later Shansabani inroads into the heart
of the Chahamana territories: In ah 588/1192–1193 ce, a garrison under the
command of the ghulam Qutb al-Din Aibek was stationed at the fort of
Kuhram, in the vicinity of Mirath and within striking distance of Delhi, and
ultimately the Chahamana capital of Ajmer (Ajayameru). See Juzjani vol.
I, 1963: 397–398, 401 (trans. vol. I, 1881: 453–454, 469–470); Edwards 2015:
96–97, 111; and Volume II.
84. There is continuing confusion between two ethnonyms, viz. Gakkhaŕ (‫)گکهر‬
and Kukhar (‫)کوکهر‬, possibly perpetuated by copyists’ orthographical errors
over the centuries. Juzjani’s mention of the last Ghaznavid sultan’s mercenary
allies as the Kukhar may have actually referred to the Gakkhaŕ. A tribal group
with the latter name has been documented in West Panjab and southern
Kashmir, characterized as a “war-like Muslim tribe … of indigenous origin
… [and] agriculturalists by profession.” The retention of some non-Islamic
practices – e.g., prohibition of widow remarriage – may indicate a gradual,
perhaps selective conversion to Islam not unlike that of the Ghuri tribes
of Afghanistan, among whom some non-Islamic practices likewise persisted
(cf. Chapters 2 and 3). The Gakkhaŕ “homeland” fell within the areas of
Ghaznavid control during the late twelfth century, though whether the tribe
was consistently agriculturalist during the last millennium remains unknown.
Cf. Juzjani vol. I, 1963: e.g., 398 (trans. vol. I, 1881: e.g., 455); Rahman 1979:
31–32; Ansari 2012; Hanifi 2012; O’Neal 2013: 65 and n. 40.
85. Juzjani vol. I, 1963: 398 (trans. vol. I, 1881: 456–457) See also Bosworth [1977]
1992: 129–131 for a more detailed account, supplemented with the reports of
historical authors in addition to Juzjani.

CHAPTER 2
Beginnings
Origins are a seductive necessity for historians.

M. A. Asif
T he epigraph above comes from M. A. Asif’s recent book on the

thirteenth-century text known as Chach-nama.1 The text was the
work of ‘Ali Kufi, dated to ah 623/1226 ce and dedicated to Sultan Nasir
al-Din Qabacha, ruler of Uchh and Multan in the southwestern Panjab
ah 599–625/1203–1028 ce. Chach-nama and its patron have some bearing
on our considerations here: they provide a preview of the historical arc
of the Shansabanis, as they emerged from obscure beginnings in central
Afghanistan and went on to create an extremely consequential imperial
expanse. The select military entourage of the Shansabani sultan Mu‘izz
al-Din of Ghazna (r. ah 568–602/1173–1206 ce) included Qabacha, who
had been a ghulam participating in the sultan’s eastward campaigns to the
Indus and ultimately to the north Indian plains during the late 1170s–1190s
(cf. Chapters 1 and 5).2 Mu‘izz al-Din was assassinated near Lahore in ah
602/1206 ce; since he had no male progeny, his closest ghulams were his
heirs, and the “patrimony” of the sultan’s conquests was divided among
his favorites, including Qabacha.3
Asif’s at times polemical position “against origins” or beginnings was,
arguably, necessary to crack the proverbial nut that has been Chach-nama
in the modern historiography on Islam in South Asia.4 Certainly, Kufi’s
own avowal that his work was an early thirteenth-century Persian transla-
tion of a lost eighth-century Arabic original5 already signaled the bifurca-
tions between historical phenomena and their phenomenologies. Whether
translation or new composition,6 Kufi’s Chach-nama was as much about
the eighth-century integration of Islam into South Asia’s cultural fabric,
as it was about the historical consequences and intentional power of these
events for its author and his audiences of the thirteenth century and
beyond. And yet more than a millennium later, the obsessive scholarly and
political pursuit of Islam’s beginnings in South Asia – often amounting to
nothing more than attestations of the Otherness of Islam and Muslims –
has had dramatic and irreversible impacts on the lives of generations of
people in this part of the world, and in South Asian diasporas elsewhere.7

80 Iran to India
Nevertheless, far from pinpointing an exact chronological or geograph-

ical point of origin per se for the Shansabanis – something that is not pos-
sible given the current state of knowledge – this chapter underscores the
fact that the origins of many protagonists throughout history, particularly
those with non-elite beginnings, remain undocumented and ultimately
unrecoverable.8 Their ethnogeneses, ensuing either from the protagonists’
own invention or a posteriori piecing together by modern academics (as
is the case here and, really, in all modern studies), are nothing more
nor less than the synthesized results of the needs and conventions of the
moment of writing inhabited by the court chronicler or modern investi-
gator. The resulting narrative may withstand sustained scrutiny or it may
not, depending on further recovery of information on the region and its
inhabitants over time. An exploration of the Shansabanis’ pre-imperial
history essentially provides a case study of the intriguing process by which
an obscure kinship-based group, practicing seasonal transhumance and/or
some form of nomadism, underwent a seemingly sudden rise to regional
and transregional prominence.
Indeed, unearthing a plausible, non-mythical early history of the
Shansabanis, and the way of life of their forebears during the eleventh
through the first half of the twelfth centuries is more than a “seductive
necessity”: it serves one of the fundamental aims of this project, namely,
“essay[ing] non-sectarian histories of conquest … conflict,” and cultural
transformation.9 The search for Shansabani beginnings helps us to con-
front the complexities inherent in “Islam,” its many cultural, political,
and religious implications, as well as the impetuses for conversion and/
or public confessionalism. Rather than a general explication of historical
adoptions of Islam, the early Shansabanis provide one concrete iteration of
what were multifarious and contingent processes across historical Eurasia,
each deserving of close analysis in its own right.10
The general paucity of textual sources, and the literary conventions
reigning in those that do exist,11 urge the investigator to turn to other
potential repositories of information, wherein there may be indications of
the Shansabanis’ pre-imperial beginnings. Among the notable alternative
avenues for gleaning clues are the available numismatic data12 and – the
focus here – the architectural remains scattered throughout Ghur (e.g.
Figures 2.1, 2.5–2.10, 2.13–2.16). With the exceptions of the Saghar minar
(Figure 2.17) and probable summer “capital” of Jam-Firuzkuh (Figure
2.2) – discussed further in the current and following chapters – the vast
majority of historical Ghur’s built environment can be characterized as
one of monumental decadence: the architectural landscape of central
through southern Ghur consisted of pre-existing square or round multi-
story towers, fortified enclosures of varying sizes on plains and promon-
tories, and possibly interconnected lookout structures along the principal
long-distance routes of travel (Figures 2.5–2.8).

Beginnings 81
Figure 2.1 Ahangaran, Ghur Province, Afghanistan. © Photograph Warwick Ball,

c. 1970s.
This chapter argues that, during the later seventh–tenth centuries, the
structures of central–southern Ghur (subsequently ruined) had surely
been erected as monuments commanding the surrounding landscape, and
served imperial purposes for the late Sasanian–Hephthalite or Western
Turk elites.13 By the eleventh–twelfth centuries, however, these same struc-
tures were no longer monuments projecting imperial presence; rather, they
were repurposed as temporary shelters for people, their animals and goods,
and likely also reused on occasion as fortifications against aggressors. These
remains were decidedly not the later imperial Shansabanis’ magnificent
madrasas, mosques, and tombs at Chisht, Jam-Firuzkuh, and Herat (see
Chapters 3 and 4, and Volume II), which housed sacral institutions and
represented a ruling presence. Nevertheless, the abandoned and decaying
structures were eminently useful for the pre-imperial Shansabanis and
other Ghuris.
Many of Ghur’s architectural remains have been surveyed, documented,
and interpreted before.14 Given the dramatically diminished direct access
to Afghanistan during the last four decades, previous studies continue to
be indispensable, essentially enabling the work of subsequent generations
of scholars from several disciplines.
I argue in a later section of this chapter that most of the previous
architectural analyses of Ghur’s architectural landscape put forward con-
clusions based on untested assumptions of chronology. Here, bringing to
the fore the tried and true art-historical methods of rigorous visual analysis
and comparative contextualization, I conduct more thorough stylistic and

82 Iran to India
iconographic examinations of these significant physical traces, comparing

them with more firmly dated examples. This fresh perspective does much
to reveal the world of the pre-imperial Shansabanis and other Ghuris,
and the effects of this early existence on the Shansabanis’ later imperial
identities.
For our purposes of attaining a glimpse of the geographical and cul-
tural beginnings of the Shansabanis during the eleventh–mid-twelfth
centuries – that is, prior to and around the time of their emergence
into the “history” conveyed by the textual sources – it is important to
focus on their principal geographic vectors of activity within present-day
Afghanistan. These vectors were of course initially concentrated in the
heartland of Ghur, geographically delimited and architecturally analyzed
below. But, as described in Chapter 1, it is worth bearing in mind that
by the 1150s, the Bamiyan valley to the east was the site of an established
collateral Shansabani lineage, while the Firuzkuh Shansabanis had also
pushed far southward to Tiginabad/Old Qandahar, beyond what were
likely their ancestral over-wintering areas in southern Ghur-Zamindawar
(see Chapter 3). Then, by the later 1160s, Shansabani presence was docu-
mentable northwest of historical Ghur in the region of Maimana (Badghis
Province) and northeastward toward Balkh, firmly within the historical
geography of greater Khurasan. By the 1180s, Shansabani control had
extended westward from Ghur to include Herat and its vicinity. All of
these expansions and their architectural footprints will be discussed in the
following chapters.
Ghur
In current political maps, historical Ghur (map, p. xvi) is surrounded by

the contiguous provinces of Farah, Herat, and Badghis on its western
flank; Faryab and Sar-i Pul to the north; Bamiyan, Daykundi, and
Uruzgan along the eastern flank; and Farah and Helmand provinces to
the south. Ghur Province itself comprises ten districts named according
to their more prominent towns. These districts appear to be much more
spacious and less numerous than the tiny and multiplied divisions of the
eastern provinces of Kabul, Logar, and Nangarhar, for example – no doubt
reflecting the great differences in population density among these regions
of the country in modern times.
This seemingly ordered landscape of the present varied greatly in the
past. The deeply carved terrain in much of Ghur makes satellite data dif-
ficult to use without on-the-ground verification.15 And the interventions
in the built environment resulting from the Mongol campaigns of the
second and third quarters of the thirteenth century may have erased many
indices of populations inhabiting Ghur’s mountainous lands, further

Beginnings 83
obscuring the beginnings of the Shansabanis’ coalescence as an imperial

lineage.16 Expectedly, then, an overall socio-cultural apprehension of his-
torical Ghur, parallel with the eastern and western reaches of Afghanistan
described elsewhere in this volume, still has not come into view. While the
tracts from Kabul westward until Bamiyan and its valleys are better known
thanks to the abundance of Buddhist monastic establishments and their
attraction for pilgrims from afar who were prolific travel writers,17 Ghur
itself has resisted such knowledge economies, its mountainous terrain and
the supposed insularity of its inhabitants making methodical exploration
more perilous and reportage less frequent.18
Despite the challenges – and perhaps because of them – a piecing
together of the region’s historical, cultural, and religious traces should be
attempted. The varying degrees of artificiality in determining the political
boundaries of nation-states in the twentieth century have already been
subjected to rigorous analysis. Such a perspective is equally necessary for
the domestic parceling within nations of provinces and states, which are
frequently delineated for ease of administration – including gathering of
census data, taxation, and revenue collection – rather than strictly adher-
ing to the historic extents of linguistic and cultural geographies. Static
administrative boundaries of modern times call for deconstruction all the
more when treating a landscape largely determined by the varying degrees
and types of transhumance and/or nomadic pastoralism among its popu-
lations. In addition to sparse references in texts, information from archae-
ological surveys/excavations and also modern observations can be useful in
better imagining the extent and geography of historical Ghur. Lastly, we
must bear in mind that the conceptualization of Ghur during the tenth
through mid-twelfth centuries not only differs from the political maps of
today, it has varied through time as well, depending on the shifting bases
of power of prominent chieftains in the region.
Ahmad Ali Kohzad,19 among Ghur’s most thorough explorers in the
mid-twentieth century, described the region as “a fortress surrounded by
yet other mountains … which has no doors, or if there be any, are narrow
valleys difficult of access.” However, such a description should not render
Ghur impenetrable, given the region’s many (principally non-sedentist?)
inhabitants, both in the past and in the present (Figures 2.11 and 2.12).
The eminent Afghanistan scholar Louis Dupree’s observations are also
worth recalling: the mountains “never truly served as barriers to cultural,
economic, or political penetration, but merely funneled peoples and ideas
along certain routes.”20 The northern reaches of Ghur, then, were hemmed
by mountains (primarily the foothills of the Hindu Kush), which, rather
than preventing movement, diverted it to the seasonally passable routes of
access to Ghur’s valleys. The region’s southern reaches bordered the dry
and flat plains of Zamindawar and Sistan (map, p. xvi).21
A constant of Ghur’s landscape would have been its major northern

84 Iran to India
river, the west-flowing Hari Rud. Although being a source for water, the
river was not easily navigated or forded, and thus did not facilitate travel
and communication. Nevertheless, its banks hosted at least two significant
locales between modern Chaghcharan and Chisht-i Sharif, associated with
the Shansabanis and their ancestors: Ahangaran lay on the river’s southern
bank, while the settlement of Jam-Firuzkuh straddled both banks (Figure
2.2). Each site was mentioned in Ghaznavid and later textual sources,
respectively. While Ahangaran (see below) served as the seat of power
of one or more of the local chieftains thought to be the forebears of the
Shansabani sultans, the latter is now generally accepted to have been the
summer capital – more likely a sard-sir royal encampment, as discussed
further below and in Chapter 3 – of the initial branch of the Shansabanis
of Firuzkuh (c. 1145 ce). These settlements and their respective roles for the
Shansabanis and their predecessors are treated in succession.
The early fort and settlement of Ahangaran (Figures 2.1 and 2.3) was
the site of the encounter between the Ghaznavid sultan Mahmud and
the Ghuri chief Muhammad ibn Suri (d. 1011), referenced in Chapter
1. This chief evidently held some sway over the sub-region of Mandesh,
which extended from the Hari Rud’s southern banks northward to the
borders of Darmashan, identified as another sub-region of Ghur. Thus,
even though Ahangaran appeared to be the northernmost fortress of Ghur
in pre-modern times, the conceptual frontiers of the region lay farther
northeast and northwest, encompassing, respectively, the Chaghcharan
basin through the edges of Gharjistan.22 Southward, Mandesh stretched to
the very heartland of Ghur, which was anchored by the mountain peak of
Chehel Abdal. Kohzad (1954a: 23) noted in the mid-twentieth century that
the area around Chehel Abdal and the Mandesh region overall “ha[d] all
the necessities for animal rearing and husbandry,” implying that subsist-
ence in the region was quite possible.23
Modern documentation at the historical site of Ahangaran (Figures 2.1
and 2.3), a fortification with two concentric walls atop a rocky promontory
rising gently above the Hari Rud, shows cultivated fields below stretch-
ing to the river banks and relying on it for irrigation.24 Notwithstanding
changes in the region’s geo-climatic features over the centuries, it is con-
ceivable that historically also, Ahangaran and its vicinity were home to
largely transhumant populations. These groups resided in constructed
villages – such as Ahangaran itself – only in the colder winter months, and
moved into tents on the banks of the Hari Rud during the warmer periods
of the year.25 In modern times, specifically the mid-twentieth century, the
well-watered Ahangaran Valley’s inhabitants were not only agricultural-
ists, some among them also wove rugs. The valley in fact witnessed the
constant visits of nomadic populations, many traveling to the month-long
Khirgai market near Chagcharan, which was “a yearly meeting of nomads
from [Afghanistan’s] remotest corners.” There, brisk trade occurred in

Figure 2.2 Jam-Firuzkuh, minar and confluence of Hari Rud and Jam Rud. © Courtesy of David Thomas.
14/09/21 5:38 PM
86 Iran to India
Figure 2.3 Ahangaran, surroundings and banks of Hari Rud. © Photograph Jawan
Shir Rasikh, c. 2010.
livestock as well as hand-made objects and textiles.26 The wide spectrum

of lifestyles within the Ahangaran Valley, certainly in the present day
and probably also in the past, reiterates the conceptual importance of the
nomad–urban continuum for understanding not only the Shansabanis’
forebears and their milieu of origin but also, as we shall see below, their
later imperial history.
By the mid-twelfth century (c. ah 539/1145 ce), the Hari Rud boasted a
second Shansabani-affiliated site at its confluence with the Jam Rud. The
probable Shansabani summer “capital” of Jam-Firuzkuh (Figure 2.2) was
established quite possibly in the vicinity of a primarily Jewish settlement
already existing nearby as of the early eleventh century.27 The site received
notable architectural patronage probably upon its inception, which con-
tinued through ah 565–575/1170s ce and perhaps later (see esp. Chapters
3–5). Farther west of Jam-Firuzkuh lay Chisht and Awbeh, the last two
being eastern satellites within the orbit of the great Khurasani emporium
of Herat. Ghur’s eastern flank beyond Chagcharan already touched upon
the mountainous edges of Bamiyan and its many valleys. Thus, north-
ern Mandesh and especially Darmashan – edging Badghis, Faryab, and
Sar-i Pul provinces on modern maps – most likely formed the northern
arc of historical Ghur as understood by the Shansabanis’ eleventh- and
twelfth-century forebears.
According to Juzjani, even after the Shansabanis’ emergence as a prom-
inent regional power in the mid-twelfth century, at least some of the elites
and their entourages and other followers continued to undertake annual,
seasonally determined migrations along a northeast–southwesterly axis. The

Beginnings 87
Shansabani elites spent only the summer months of the year at Firuzkuh,
despite this “capital” being monumentally conceived, as noted above. It
was approximately forty farsangs to the south in the region of Zamindawar,
and probably at a specific location therein, that the Shansabanis spent
the winter months.28 Firuzkuh, then, was essentially the Shansabanis’
sard-sir or summer encampment of their seasonally transhumant existence,
complemented by the garm-sir or over-wintering area in the region of
Zamindawar near the borders of Sistan.29 For our present purposes, these
predictable movements furnish the broad regional span of the pre-imperial
Shansabanis during the eleventh–twelfth centuries.
Ahangaran and Jam-Firuzkuh (Figures 2.1 and 2.2) merit juxtaposition
not only for their locations along the Hari Rud, but also for their varying
suitability as the Shansabanis’ eventual sard-sir royal encampment. Indeed,
scholarly debate as to whether Jam-Firuzkuh was in fact the Shansabani
summer “capital” endured through much of the twentieth century, cen-
tering precisely on the site’s unsuitability. Aside from the admittedly mag-
nificent minar (Figure 2.2) and even with the archaeological recovery of its
adjacent, monumental mosque (cf. Chapters 3 and 4), little else indicates
the site’s royal status. Furthermore, its location was remote, not easily
accessible along the northern routes from Herat toward Balkh; and its fast-
ness in a mountainous landscape provided limited space for pasturage, also
being insufficient for seasonal agriculture along with the encampments
of royal retainers and soldiery. By contrast, Ahangaran (Figures 2.1 and
2.3) met most of these requirements, with its location on a well-watered
plain. This locality had the added advantages of being one of the crossing
points of the Hari Rud, as well as having ancestral associations for the
pre-imperial Shansabanis.30 Nevertheless, little doubt now remains that
the Shansabanis chose Jam-Firuzkuh over Ahangaran as the site for their
summer residence.
Beyond the above-mentioned migration patterns, little precision is
forthcoming in delineating the southern extent of Ghur. Zamindawar as a
historical region has been identified, of course, but in the current state of
scholarship a specific locality as the likely site of the Shansabanis’ winter
“capital” can be only tentatively suggested. Calculations of its possible
location have been put forth, though inconclusively, based on Juzjani’s
mentions of distance from Firuzkuh and comparison with the numerous
fortifications identified in the southern part of Ghur, particularly in the
area of Taiwara and perhaps beyond31 (then extending into modern Farah
province), as discussed below. Southern Ghur was evidently well-watered
by its three rivers: the long Farah Rud, the Rud-i Ghur, and the Khwash
Rud. By the twelfth century, then, it was this c. 250-km expanse from
northern Ghur to the borders of historical Zamindawar and Sistan that
constituted the core home region of the eventual Shansabani elites and
their seasonal transhumance.

88 Iran to India
Figure 2.4 Detail of Herberg’s fieldwork in central and southern Ghur, c. 1970s.
From Herberg 1982: II.
Aside from Ahangaran and Jam-Firuzkuh, many other architectural

remains dot historical Ghur’s central–southern stretches (Figures 2.5–2.11),
being generally scattered with few discernible areas of concentration.
The photographic documentation in Ghur in the 1960s by the intrepid
traveler-photographer Josephine Powell (mentioned in this volume’s
Introduction), and architectural surveys by the scholars Werner Herberg
and Warwick Ball in the 1960s and 1970s,32 reveal a cultivable but rugged
landscape. In Herberg’s (Figure 2.4) survey of the structures still standing
at various locations spanning from north of the Farah Rud through the
southern areas beyond the Rud-i Ghur, he observed that in general they
rarely stood atop peaks or inclines, but commonly at the bases of cliffs
and protectively flanking passages through valleys (Figure 2.5).33 At times
multiple structures were also grouped together, so that they could have
enclosed (partly or completely) cultivable areas or pastures (Figure 2.6).

Beginnings 89
Figure 2.5 Yaman, Ghur Province, Afghanistan, ruins of a square tower and
cruciform “military structure” with remains of defensive wall. From Herberg
1982: fig. 3.
Figure 2.6 Ghor Province, Yahan, fortification line, ruins. © Photo by


90 Iran to India
Since the vast majority of the surviving structures in the region appear
to have been intended for defense and surveillance, scholars have inter-
preted it as the stage for continual skirmishes and close-quarter battles.
Such a picture of historical Ghur was first painted by Bosworth (see also
Chapter 1), viz. an isolated and politically fragmented region of “forti-
fied domesticity” whose internally squabbling inhabitants repelled out-
siders throughout historical memory.34 Furthermore, modern studies
have accepted at face value Juzjani’s assertion that ‘Abbas ibn Shish
(early–mid-eleventh century), a scion of the rival Shishani lineage (also
collateral to the pre-imperial Shansabanis), built the fortifications still
standing in central–southern Ghur, thereby dating them to a short period
probably sometime in the first half of the eleventh century.35 Indeed,
this domino effect of accepted assumptions appears to have resulted in a
focus largely on defensive structures in southern Ghur. Ball, for example,
explained an apparent concentration of defense and surveillance struc-
tures in Ghur’s southern reaches as the Shansabanis’ principal line of
defense against the Ghaznavids, necessitated by the latter’s vast empire and
particularly their presence in the contiguous area of Bust-Lashkari Bazar
during the eleventh century.36
However, a much-needed re-examination of central Ghur’s built envi-
ronment is undertaken here, encompassing the typologies, and decorative
styles and iconographies of the built remains themselves. It is equally
important to bring to bear upon the analysis what is known of the life-
styles and economies of the pre-imperial Shansabanis and their fellow
Ghuris. I point to the possibility of an alternative time span of construc-
tion for at least some of southern Ghur’s standing structures: the late or
post-Sasanian through early Islamic (seventh–tenth centuries) periods,37
which would thereby pre-date the Shansabanis and their immediate
forebears. Additionally, the documented traces of decoration on some of
Ghur’s architectural remains plausibly suggest non-defensive functions
for them as well. These new possibilities for understanding Ghur’s archi-
tectural landscape are not only supported by the architectural evidence,
they also fit better into the more nomadic/transhumant lifestyles of the
Shansabanis’ pre-imperial ancestors. But, as ever, it must be remembered
that the ensuing analyses and conclusions remain tentative, based on the
documentation of scholars working in the region through the late 1970s
when it was more accessible.
It was noted above that Ghur’s built remains comprise either walled
enclosures with integrated square or semi-circular corner bastions, or
individual “dwelling towers,”38 the latter being more common (Figures
2.6–2.8). The construction materials were locally sourced and consisted
primarily of sun-dried bricks, which formed walls atop stone footings
varying in height (Figure 2.9). Some variation in brick sizes was also
observed, ranging from 33 cm to 48 cm in length and 8 cm to 11 cm in

Beginnings 91
Figure 2.7 Ghor Province, Male Alau, brick tower, ruin, distant view.
© Photograph Josephine Powell c. 1960. Harvard Fine Arts Library,
height, but any regional correlations or chronological indicators have as

yet to be established. Although baked bricks were present, they were never
documented in situ, but rather scattered nearby, and may have been used
sparingly for decoration or perhaps now collapsed parts of structures.
Wood seemed to play virtually no role in the construction.39
All structures tended to be up to three stories tall (Figure 2.10); many
of the towers as well as larger complexes, consisted of superimposed rooms
within a tower, or a small number of spaces within an enclosure. They
hinted at short- or long-term habitation, with some of the interiors of
upper rooms bearing traces of possible painted decoration, indicating that
they could have been the living quarters; however, the date range of these
decorations remains unknown. These possibly residential rooms were
raised above lower ones that could have been used for storage, or perhaps
shelter for livestock.40 Since excavations have not been carried out at any
of these locations, and only surface surveys with photographic documen-
tation and schematic drawings are available, the patterns of usage remain
conjectural.

92 Iran to India
Figure 2.8 Ghor Province, Alana Valley (junction of routes to Khissar, Parchaman,
Nili, and Tiawara), fortification line, ruins, distant view. © Photograph
According to the drawings of the surviving structures made by Herberg

and Ball, there is a remarkable variation in ground plans.41 As noted above,
the towers were circular or angular on plan, and either independent or
addorsed to larger enclosures or buildings. The larger complexes exhibited
the greatest variety of plans among themselves, ranging from rectangular
enclosures to cruciform structures (e.g. Figure 2.10), with some combining
these forms in more complex layouts. While a more systematic survey and
certainly archaeological excavation could both aid in determining at least
a relative chronology or developmental sequence among these plans, at
present the documentation stands as it is, lacking the evidence for larger
patterns of change over time.
Overall, southern–central Ghur’s surviving structures seem recalci-
trantly resistant to dating, both in terms of period of construction and
longevity of use. No epigraphic data have been recovered, so there are
no known foundation dates, patrons, or builders as points of departure

Beginnings 93
Figure 2.9 Taywara vicinity, Ghur Province, watch tower. From Herberg
1982: fig. 1.
for plausible spans of occupation/use, or clues regarding the functions

of the sites and structures. Nor does the fabric of the remains them-
selves, comprising walls primarily of unbaked brick (rarely baked) atop
tall stone footings – as described above – provide diagnostics of date, as
this construction method was prevalent in pre-Islamic strata throughout
Afghanistan.42 In the face of these challenges, it is worth interrogating –
insofar as this is feasible, with parallel or illustrative data supporting the
little direct evidence – whether the pre-imperial Shansabanis and other
Ghuris of the eleventh century were indeed the fashioners of this land-
scape, as implied by Juzjani and generally accepted in modern scholarship.
Would these groups’ historical habitus, largely constituted of certain life-
styles and economies, have allowed or even called for the construction of
numerous “dwelling towers,” fortified enclosures, and other structures for
defense and surveillance?

94 Iran to India
Figure 2.10 Between Pasa Band and Dahane Nawrak, cruciform interior of a
defensive structure. From Herberg 1982: fig. 6.
As described in the preceding chapters, the early Shansabanis and other

Ghuris spanned the continuum between nomadism and sedentism, with
many groups engaging in seasonal transhumance combined with spot
agriculture. Modern Taimani and Firozkohi nomads inhabiting the areas
immediately north and south of the Hari Rud (east of Herat) provide illus-
trative parallels in broad contours,43 which can inform historical under-
standings of transhumant and nomadic–pastoralist groups such as the
pre-imperial Shansabanis and their fellow Ghuris.
The southern Taimanis’ summer shelters of rectangular black tents
(Figure 2.11), made of willow poles and woven goat hair coverings (palas),
required no specialized architectural skill per se. Although it is notewor-
thy that their tents tend to mimic winter village dwellings, and the tent
coverings are traditionally woven by women. By contrast, the Firozkohis’
yurts (Figure 2.12) were heavier and laborious to erect, resulting in a more
expensive dwelling that also required camel or other large-scale trans-
port. Given the yurt’s characteristics as a dwelling type, it was used by
groups with short migration circuits, and only in regions within the
circulatory patterns of artisans skilled in crafting their specialized wood
frames.44
All in all, the lifeways described above appear to preclude the need for
coordinated projects of immovable architecture,45 such as dwelling towers
and fortifications interconnecting vast swaths of land. Transhumant and
nomadic–pastoralist existences instead require principal concentrations
of time, energy and resources upon moveable dwellings, animal hus-

Beginnings 95
Figure 2.11 Taiwara-Farah Rud. © Photograph Warwick Ball, c. 1970s.
Figure 2.12 Ahangaran. © Photograph Warwick Ball, c. 1970s.
bandry, small-scale farming, and the production of portable commodi-

ties. Transhumant rather than nomadic segments of the population – for
example, the inhabitants of Ahangaran in northern Ghur, of whom the
above-mentioned Firozkohis might be descendants – could have estab-
lished settlements for seasonal occupation during specific periods of the
year (Figures 2.1 and 2.3). Moreover, the specialized expertise required in
erecting relatively portable winter dwellings, such as yurts, appears not to
be endemic to transhumant groups, but rather the latter purchase it from
skilled crafters, and that only where these workers are available. In fact –
and insofar as modern praxes reflect historical ones – even these artisans’

96 Iran to India
specific expertise of fashioning construction elements in wood would have

evidently had no place in the exclusively brick and plaster architecture of
the region (see above). Southern–central Ghur’s building programs, then,
seem altogether dissonant with mainly transhumant and nomadic ways of
life.46 Rather than constructing entirely new and compex built environ-
ments, the region’s historical inhabitants’ varying degrees of movement
would instead have called for repairing, repurposing, and reusing inherited
architectural landscapes.
During A. A. Kohzad’s tours of central–southern Ghur in the
mid-twentieth century, his local informants among the region’s inhab-
itants, particularly in the Yaman Valley southeast of Taiwara, assigned
notably different builders to the towers/fortifications, attributing virtually
all the local ruins to the pre-Islamic centuries.47 There may be a basis for
these oral histories: one of the most noteworthy features of many of the
region’s ruined structures, whether sole towers or larger complexes, is
their surface decoration (Figures 2.13–2.16). Based on first-hand docu-
Figure 2.13 Ghor Province, Yahan, fortification tower, ruin, with decoration
typical of region. © Photo by Josephine Powell c. 1960. Harvard Fine Arts Library,

Beginnings 97
Figure 2.14 Southern Ghur, southeast façade of a structure. From Herberg

1982: fig. 9.
mentation, Herberg described these structures as thickly plastered on the

exterior, with many having decoration incised or molded on their surfaces
while the plaster was still wet.48
The decorative motifs on central–southern Ghur’s remains evince some

98 Iran to India
Figure 2.15 Qala‘-yi Chahar Baradar, Ghur Province. © Photograph

Warwick Ball, c. 1970s.
consistency, being principally shallow patterned hollows creating trian-

gle and diamond shapes across much of the surface (Figures 2.13–2.16).
Motifs of arches in series were confined toward the tops of structures
(Figures 2.16 and 2.23). In some cases, such as the walls of Qala‘-yi Chahar
Baradar and the stone-footed brick towers at Muna Ala (Figures 2.15 and
2.16) – both within a few kilometers of each other – square or rectangu-
lar panels framed tightly wound and molded circular patterns evoking
stylized flowers or floral motifs, interlaced with gently curving double
lines.49 The hemispherical arch series toward the top of the Muna Ala
tower (Figures 2.16 and 2.23) has additional molded decorations within
the arches, possibly derived from a shell or floral design (see below). What
appear to be column capitals at the arches’ springing were emphasized with
a triangular motif in relief. Above the arches more horizontal, frieze-like
panels containing thicker circles with smaller central holes, interlaced with
the diagonal lines are visible. Again, it may be tempting to consider these
decorative motifs as diagnostic of at least a relative chronology in which
to place their towers or structures, for example, one “progressing” from
simple to complex in order to explain the varying intricacy and variety
of decoration. However, any timelines must remain tentative, given the
absence of supporting archaeological and epigraphical data.
The method of fashioning the decoration found on many of central–
southern Ghur’s ruins, as well as its iconography and style (Figures 2.13–
2.15), all stand in notable contrast to the known Saljuq and Ghaznavid
minars that have been brought up as comparisons (late eleventh–twelfth
centuries; cf. also Chapter 1). More significant for our purposes, they diverge

Beginnings 99
Figure 2.16 Ghor Province, Male Alau, brick tower with stone base, ruin, with
scroll pattern decoration. © Photograph Josephine Powell c. 1960. Harvard Fine
from the Da Qadi minar (Figure 2.17): this recent rediscovery has been put
forth as possibly the early Shansabani-patronized “lost” minar of Qala‘-yi
Zarmurgh, located near Saghar, 75 km southwest of Jam-Firuzkuh.50
Surface deterioration has been extensive on the Da Qadi tower, leaving
behind no inscriptional traces and lacking the bottom bands of decoration
and epigraphy horizontally framing the shafts of most twelfth-century
Saljuq minars. Nonetheless, its tapering cylindrical shaft, the remaining
upper bands of geometric decoration, and the division of the minar’s shaft
by means of these and likely additional bands, overall emulate the region’s
better-known minars, such as the no longer extant one at Qasimabad in
Sistan (c. 1125–1150 ce) (Figure 2.18) and the Daulatabad minar south of
Balkh (early twelfth century) (Figure 2.19).51
Although the minar at Da Qadi will be analyzed in greater detail
later in the chapter, it should be noted here that the ornamentation on
central–southern Ghur’s structures (Figures 2.13–2.16, 2.20, 2.23) differs
markedly from the Da Qadi minar as well as the other minars (Figures

100 Iran to India
2.17–2.19) in material as well as form:

the latter’s decorative bands were
fashioned with baked brick and some
stucco so that, where brick was used,
the decorative repertory was limited
to geometric and minimally curving
designs.52 By contrast, the exclusive
use of a malleable plaster on Ghur’s
central–southern remains accom-
modated curving and semi-floral
patterns, even in deep relief. While
the Da Qadi minar emulated the
baked-brick Saljuq and Ghaznavid
towers – aspiring to a “cosmopol-
itan” building tradition prevalent
throughout the Persianate regions, as
discussed below – central–southern
Ghur’s qasr, qala‘ and kushk emerged
from a distinct architectural culture.
Despite Juzjani’s general attribution
of the ruins to the tenth or early elev-
Figure 2.17 The “lost” minar of Qala‘-yi Zarmurgh, enth century, and modern scholars’
c. 75 km south of Jam-Firuzkuh. From Thomas general acceptance of the date, many
2014: 26.
of central–southern Ghur’s ruins
point toward an earlier span of time,
namely the late or post-Sasanian–early Islamic periods of the late seventh–
tenth centuries. (It is impossible to ascribe this timeframe to all of Ghur’s
remains, given the current inaccessibility of the region for first-hand
documentation.)
The pre-Islamic affinities and chronology of the construction and deco-
ration of many of central–southern Ghur’s structures are borne out by jux-
taposition with stucco architectural decoration dating to the later Sasanian
centuries (Figures 2.21, 2.22, 2.24). Several sites in the vast empire’s east-
ernmost reaches – notably excluding western Afghanistan53 – have under-
gone archaeological exploration/excavation and analyses of their stucco
decoration, viz. Bandian (fifth century ce), Tepe Hissar near Damghan
(fifth–sixth century), and Mele Hairam near Sarakhs (fifth–seventh cen-
turies) (all in modern Iran). However, given the fragility of stucco as a
material, and the consequent loss of much of the original Sasanian corpus,
relevant examples from different periods and regions of Sasanian ascend-
ancy must also be considered.54
Repetitive and abundant molded patterns suffused many of the plaster
surfaces of central–southern Ghur’s built remains (Figures 2.13–2.16, 2.20),
demonstrating a palpable resonance with Sasanian parallels. The curved and

Beginnings 101
Figure 2.18 Minar of Qasimabad, Sistan. From Tate 1912–1912: opp. p. 22.
at times tightly wound double lines on Ghur’s buildings (Figure 2.20) –

often framing or meandering among flush, deeply carved or stamped floral
motifs – echo the widespread framing technique of Sasanian stucco dec-
oration: relevant examples come from both the Sasanian east at Tepe
Hissar (Damghan, fifth–sixth century ce) (Figure 2.21), as well as the west
at Nizamabad (south of Tabriz) (Figure 2.22).55 Moreover, the stylized,
high-relief rendition of floral elements characterizing the known Sasanian
stucco corpus forms an intriguing continuum with the fleshily sculpted
floral whorls on some of Ghur’s structures (Figures 2.15, 2.16, 2.20, 2.23).

102 Iran to India
Figure 2.19 Balkh, minaret of Daulatabad (1108–1109), exterior, full view:

cylindrical, decorated in “light and dark” (Hazar Baf) style of brickwork; bands of
Kufic and Naskhi script, both inscriptions and decoration; bricks, unglazed, and
incised stucco. © Photograph Josephine Powell c. 1960. Harvard Fine Arts Library,
Figure 2.20 Detail of Figure 2.16, Male Alau, Ghor Province, brick tower,
lower decoration. © After Josephine Powell c. 1960. Harvard Fine Arts Library,

Beginnings 103
Figure 2.21 Damghan, Semnan Province, Iran, stucco dado with palmettes,
Sasanian. From Kröger 1982: tafel 89.
Motifs of arches in series on the plastered surfaces of Ghur’s remains –

for example, toward the top of the Muna Ala tower (Figure 2.23), also
discussed above – illustrate the capacious heuristic of architectural culture
and the nature of the relationship between the Sasanian expanse’s “cos-
mopolitan” stucco decoration, and the manifestation of these practices in
the empire’s eastern reaches in modern Afghanistan. These arcade motifs
were also part of the iconography of Sasanian stucco decoration, as seen
at sites such as Taq-i Bustan (c. fourth century ce) (Figure 2.24), where
the finely worked stucco covering a column capital exhibited arches in
series with exquisite, seashell-like striations in their semi-circular hoods.
This common decorative motif was, as is evident, adapted into the local
repertory of south–central Ghur as well, but here the fine, shell-inspired
lines were interpreted as a broad three-lobed design – likely the outcome
of the differing pliancy of the available plaster and the skill of the artisans
working it.

104 Iran to India
Figure 2.22 Nizamabad, Iran, wall fragment with stucco decoration, Sasanian/early Islamic?
From Kröger 1982.
Figure 2.23 Detail of Figure 2.16: Male Alau, Ghor Province, brick tower,
upper decoration. © After Josephine Powell c. 1960. Harvard Fine Arts Library,
The proposal of a late seventh–tenth century span for the initial construc-
tion of central–southern Ghur’s architectural remains – their renovation,
repurposing, and reuse possibly continuing well into the twelfth century
and beyond – also concords with the region’s historical trajectory. Given
the significance of Sistan–Khurasan for both Sasanian imperial consoli-
dation, and the later Sasanians’ lingering rebellions against the onslaught
of Arab-Umayyad ascendancy (see below), the presence of pre-Islamic

Beginnings 105
Figure 2.24 Taq-i Bustan, central Iran, column capital (Sasanian). Kröger 1982:
tafel 40.
remains in the adjacent parts of Ghur should not elicit surprise. Indeed,
the Islamic-Umayyad and -Abbasid extensions into Sistan during the late
seventh through eighth centuries did not progress unchecked, but rather
were protracted over at least the next 150 years. Leaders with local power
bases, such as the Saffarids (eighth–thirteenth centuries, with various
suzerains), themselves arising from the earlier influx of Umayyad and
Abbasid expeditionary forces, embarked on their own ongoing struggles
for supremacy.56
The historical sources describe incessant confrontations between the
Rutbils and – initially – the Umayyad- and Abbasid-deputed forces, and
eventually Saffarid loyalists and others. “Rutbil” was likely a hereditary
title for the rulers of Zabulistan, parts of al-Rukhkhaj, Zamindawar,
and Sistan,57 whose seasonal capitals were distributed between northern
sard-sir or summering grounds in Zabul, and the southern garm-sir in
Zamindawar in the winter months (map, p. xvi).58 The military engage-
ments between the enigmatic Rutbil’s allied military contingents (includ-
ing “Turks”) and the equally varied, Islam-affiliated forces were frequent
and ongoing at least up to the end of the ninth and likely into the tenth

106 Iran to India
century. According to Tarikh-i Sistan, the bellicosity was prolonged as

most confrontations between the military presences did not result in defin-
itive eradication of either side: in cases of victory for the Islam-affiliated
forces, the Rutbil paid high indemnities in lieu of being deposed outright,
which in turn perpetuated general instability in the region.59
Equally significant for our purposes is a discernible pattern among
strengthening or newly coalescing polities in the aftermath of the Sasanian
empire’s dénouement: their emulation of the Sasanians through architec-
tural patronage and continuity in ceremonial and ritual cultures, essen-
tially “assert[ing] their legitimacy by placing themselves firmly in the
line of past Iranian kings”.60 The sandwiching of southern Ghur between
Zamindawar and Sistan(-Khurasan) could have facilitated the extension
there of a “Sasanianizing” architectural culture during the late centuries of
the first millennium ce.
The working conclusion I offer, then, is that Ghur’s already existing
towers and fortified structures continued to play an important role for the
Shansabanis’ pre-imperial ancestors as well as other Ghuris, who formed
part of the complex and ever-changing nomad–urban continuum of Ghur
and its contiguous regions. The Ghuris – including among them the group
who would eventually identify themselves as the Shansabanis – repurposed
and/or reused this inherited landscape and its built environment, especially
during their winter sojourns in the garm-sir of southern Ghur, touching
on the borders of Zamindawar and Sistan. In light of the alternative dating
proposed here for central–southern Ghur’s towers and other structures,
it seems worthwhile to reassess the dismissive characterization of Ghuris
in general as an unruly and decidedly disruptive people. A more nuanced
understanding of their political and cultural milieu prior to the twelfth
century is now beginning to emerge, as we bear in mind both their degrees
of transhumance/nomadism, and their inheritance of a built environment
largely pre-determined by their Hephthalite, Sasanian, and/or Western
Turk predecessors.
During the eleventh and twelfth centuries, Ghur’s location as “the
buffer-zone between [the] two empires” of the Ghaznavids and Saljuqs
apparently ensnared the region within the imperial wrangling and skir-
mishes often taking place uncomfortably close to its permeable “borders.”61
While this point of contact was admittedly a politico-military bane for the
Ghuris, architecturally speaking it could be seen as a boon: Sasanianizing
trends had already entered the region, as noted above. Furthermore, the
pre-eminent imperial formations of the eleventh–twelfth centuries vied
for supremacy in its environs, enabling the influx of the brick-based archi-
tectural culture emerging from farther west in Saljuq Iran (cf. Chapter
1). This identifiable set of building practices incorporated some compo-
nents from their Sasanian predecessors, but essentially were coalescing
as the transregional and “cosmopolitan” imperial language of a new age,

Beginnings 107
dawning as much on the larger Persianate world as on the emerging

Shansabani clan.62
Emergence into “History”
The oft-quoted anecdote in Juzjani’s ˝abaqat of the inter-clan competition

between the Shansabanis’ mythical progenitor Amir Banji ibn Naharan
and another important personage of Ghur called Shish ibn Bahram appears
to be a legendary “origin myth” more than reportage of fact. The story’s
basic plot consists of the rivalrous Ghuri chieftains seeking mediation by
the great ‘Abbasid Caliph Harun al-Rashid (r. ah 169–193/786–809 ce)
himself, based at his magnificent court in Baghdad. Juzjani’s tale also
described the intercession on behalf of Banji by a Jewish merchant: this
figure’s function as a synecdoche of the Jewish communities of Afghanistan
has already been borne out by the epigraphical, architectural, and archae-
ological evidence gradually emerging from the region over the last century
of surveys and documentation, illuminating its socio-economic history.63
This “origin myth,” then, may hold further historical kernels, akin to the
noted confessional and professional affiliation of one of the protagonists,
viz. the Jewish merchant.
In essence, Juzjani’s story was a Pygmalion-like tale of the advantages of
transformation from provincialism to urbanity. Both the mentioned Ghuri
chiefs were rough and uncouth in dress and manner, given their origins in
a remote area supposedly out of the reach of the refinements of Islamicate
civilization – arguably at one of its apogees by the ninth century; both of
them were therefore unfit to enter the caliph’s court. But Banji was friendly
with a Jewish merchant from Ghur, cosmopolitan and wise in the ways
of the world due to his own far-reaching travels in pursuit of commerce.
Heeding his advice on dress and behavior, Banji came into the caliph’s
presence much better prepared than his rival Shish. In Harun al-Rashid’s
arbitration, Shish ibn Bahram was awarded military command over Ghur,
while Banji was awarded the region’s governing authority and the title of
amir – the latter clearly the superior position and presumably the greater
reward for Banji’s ability to adapt to the cultured ways of comportment and
overall self-presentation prevailing at the caliph’s glittering court.64
Juzjani’s pithy anecdote certainly mythologized the age-old and con-
tinuing rivalry between the Shansabanis and their collateral relations the
Shishanis, resurfacing periodically even after the two lines were allied
by marriage (see esp. Chapter 4). More compellingly, it hinted at the
Shansabanis’ origins in central Afghanistan’s nomad–urban continuum
of the early second millennium ce (continuing into modern times).
But ultimately, I propose that the story encapsulated the Shansabanis’ –
and perhaps also the Shishanis’ – own larger process of transformation

108 Iran to India
from transhumance and/or nomadic pastoralism into a more sedentist

and courtly imperial lineage, commanding (at least for a few decades)
an ostensibly Perso-Islamic empire via a seasonally transhumant elite.
The logically following lines of investigation, then, would encompass at
least an approximate timeframe in which this transformation occurred,
and the possible impetuses behind it. Moreover, how thorough was
the transformation – were pre-imperial mores relinquished completely,
or maintained to a discernible degree even as the Shansabanis attained
regional and then transregional prominence? Finally, what role did Islam
play in the process? Again, the architectural remains within historical
Ghur provide a unique point of departure for answering these questions.
Processes of socio-political transformation are virtually impossible to
recuperate from textual sources alone: as we have seen, the Shansabanis’
pre-imperial history is not thoroughly reported even by Juzjani (cf.
Chapter 1), surely in part due to their less than favorable “barbarian”
origins. At the same time, material culture primarily records these subtle
and longue durée processes’ outcomes, such as manifest alterations in the
built environment, rather than the processes themselves (cf. Introduction
and Chapter 1). However, consideration of the available traces in con-
junction with each other, and with anthropological studies document-
ing shifts in lifeways and economies within modern transhumant and/or
nomadic societies, may provide informative parallels for conceiving of the
Shansabanis’ historical transformation.
It was argued earlier in this chapter that the pre-imperial Shansabanis
and other Ghuris of the eleventh and early twelfth centuries, still prac-
titioners of varying degrees of transhumance/nomadism, repurposed or
reused Ghur’s extensive architectural remains. In the present attempt to
recreate with some plausibility their process of transition toward regional
politico-economic dominance, the recent rediscovery of a minar-like
tower at the rural settlement of Da Qadi (Figure 2.17), 75 km south-
west of Jam-Firuzkuh, may prove to be extremely fortuitous. The minar
gives credence to Juzjani’s mention of such a structure at the site of
Qala‘-yi Zarmurgh, which has been plausibly located in the vicinity of Da
Qadi. Moreover, Zarmurgh was supposedly the place of origin of the
above-described rube-become-cosmopolitan Amir Banji, the legendary
progenitor of the Shansabanis during the eighth–ninth centuries.65 This
significant rediscovery may signal and even help to date the early phases
of the Shansabanis’ public avowal of Islam – rather than their conversion
per se, as discussed below. Ultimately, it could help mark the beginning
of the Shansabanis’ ability to patronize skilled laborers with exposure
to building practices that were transregional, prevalent beyond Ghur’s
nomad–urban continuum, and which were increasingly pertinent as the
Shansabanis began to emulate their imperial contemporaries the Saljuqs
and the Ghaznavids.

Beginnings 109
Although first-hand documentation of the minar at Da Qadi must

await the region’s (and Afghanistan’s) greater accessibility, David Thomas
and his colleagues have analyzed the tower as far as possible through
photographs, combining their observations with the requisite input of
local villagers with knowledge of the structure, its surroundings, and other
historical remains in the vicinity.66 Understandably, the authors were
reluctant to assign a definitive date to the tower, owing not only to the
current impossibility of rigorous excavation or direct study at the site, but
also due to the absence of epigraphic fragments on or around the remains
that could stylistically hint at chronology. Generally speaking, Thomas
and his collaborators related the minar’s surviving decorative iconography
and style principally to Saljuq-patronized towers in the Persianate expanse
dating to the first half of the twelfth century.67
As described earlier in this chapter, the Da Qadi minar shows stark dif-
ferences with central–southern Ghur’s reused and repurposed architectural
remains, the two broad types of structures subscribing to distinct architec-
tural cultures. Ghur’s stone-footed, brick and plaster ruins hearkened to
the late Sasanian–early Islamic presences in the region during the seventh–
tenth centuries. Meanwhile, the Da Qadi minar, made of baked brick
with some additions of wood and clay for structural integrity, ensued from
very different building practices: this minar and related parallels formed
an essential typology within an architectural culture initially fomented by
Saljuq patronage,68 which eventually became the transregional and cos-
mopolitan architectural language adopted by imperial formations of the
eleventh–twelfth centuries throughout the Persianate world.
The dissemination of cosmopolitan Persianate building practices in
southern Afghanistan appears to have occurred certainly by the beginning
of the eleventh century, and possibly before. Saljuq minars surely served
as aspirational impetuses for Ghaznavid towers, the surviving examples
having been attributed to Mas‘ud III (r. 1099–1114) and Bahram Shah
(r. 1118–1152), now the most prominent remnants of Ghazna’s once mag-
nificent urbs (Figure 2.25). The Ghaznavid minars’ stellate – rather than
cylindrical – plans formally evoke the remnants of Buddhism in their vicin-
ity, as well as two known towers from central Sistan: one at Zaranj-Nad-i
Ali (Figure 2.26); the other, known as Khwaja Siyah Push (Figure 2.27),
near Chakhansur (north and northeast of Qasimabad, respectively). These
towers’ plans combined pointed and semi-circular flanges, and have also
been cited as precedents for Delhi’s Qu†b Minar (founded c.1199 ce).69
The early Shansabanis would not have had to look impossibly far, then,
for workers specialized in brick construction, whose skills would have
been required for building the minar at Da Qadi: several such towers were
already standing in the contiguous regions of Sistan (Figure 2.18)70 and
Zabulistan (specifically Ghazna) (Figure 2.25), as well as farther afield at
prominent emporia like Balkh (Figure 2.19).

110 Iran to India
Figure 2.25 Minarets of Ghazni, distant view. © Photograph Josephine Powell

In the process of assigning the Da Qadi minar a definitive place within

the twelfth-century phase of this cosmopolitan Persianate architectural
culture, it is important to note the similarities as well as the significant
differences between this tower and its relevant contemporaries, specifically
those at Ghazna and in the Sistan and Balkh regions. For it is inescapable
that the Da Qadi exemplar diverges from these others in scale, and in the
variety and quality of its decorative program.
The Ghazna minars (Figure 2.25) likely measured about 44 meters in
height when both the stellate and cylindrical shafts were in place (prior
to c. 1915; cf. Figure I.20); while the Balkh and Qasimabad towers (Figure
2.18 and 2.19) reached at least 30 meters.71 Based on the average dimen-
sions of the fired bricks used in such constructions, Thomas et al. pro-
posed that Da Qadi’s minar could have been between 7.2 and 10.4 meters
in height.72 Further, even though Da Qadi’s tower followed its larger
comparanda in tapering toward the top, perhaps also having balconies
in wood or another material, it probably did not have an upper shaft (as

Beginnings 111
Figure 2.26 Nad-i Ali, Zaranj, Sistan, remains of tower. From O’Kane 1984, fig.
14c. Photograph T. Ward (early twentieth century) from the Royal Geographical
Society, London.
at Ghazna). In sum, at approximately one-third the height of the Balkh

and Qasimabad minars – and probably only one-fourth the height of the
Ghazna towers – Da Qadi’s minar would not have reached the impressive
proportions of the works commissioned by royal and elite patrons in the
broader region.
Coupled with its comparatively diminutive scale, the Da Qadi tower
also had a restricted decorative program. All of the other minars men-
tioned above exhibit a variety of decorative iconography, ranging from
crisp geometry and Kufic epigraphy executed in brick, to dense floral
motifs and naskh inscriptions executed in stucco.73 By contrast, Da Qadi’s
minar shows only the geometric motifs resulting from the arrangement of
bricks at different angles in series. While the tower could have had a plaster
or other veneer with additional decoration, as it stands the structure shows
only the remains of brick-based decorative elements.
The not insubstantial differences in scale and decoration between Da
Qadi’s minar and the royal- or elite-patronized parallels at Ghazna and
those in the Sistan and Balkh regions – not to mention the minar of
Jam-Firuzkuh itself (see Chapters 3 and 4) – could lead to the conclusion
that the Da Qadi example belonged to a different time period, or even to
a different tradition altogether. However, Thomas and his collaborators
pointed out that these differences may “indicate limited means and/or
technological knowledge.”74 As put forth in the Introduction, an archi-
tectural culture should be conceived as a spectrum spanning elite or royal,

112 Iran to India
Figure 2.27 Remains of baked-brick tower, Khwaja Siyah Push, Sistan.

From Fischer et al. 1974: ill. 253.
“cosmopolitan” patronage as well as non-elite, “vernacular” iterations:

Rather than one being of lesser or greater artistic merit, all instantiations of
an identifiable architectural culture serve as equally valuable and informa-
tive historical expressions. For now, we must be satisfied with the proviso
that more thorough access to Da Qadi and its surroundings will eventually
provide less equivocal answers to the questions still lingering around the
date and architectural–cultural belonging of its minar.
Without epigraphical and other types of evidence, at present it is pos-
sible to put forth only hypotheses regarding the timeline and impetuses
for constructing the Da Qadi minar. But based on Juzjani’s mention of
such a structure and its location, it is not unreasonable to propose that
the tower was the result of Shansabani patronage at the beginning of the
twelfth century. In turn, we must trace the consequential ramifications of
such a proposal. The tower itself was not only an “Islamic” gesture, it was
pronouncedly an act of architectural patronage. It could thereby signal not
only the Shansabanis’ public declaration of an Islamicate identity, but also
the beginnings of an awareness of and transition toward the Persianate cul-
tures of power prevalent among the elites of this geo-cultural expanse. As
discussed in Chapter 1, these cultures were iterated distinctly within each
polity, yet they can be conceived as founded upon three broad spheres
of activity, viz. Islamic confessionalism, patronage, and mobility. Quite
conceivably, in the first quarter of the twelfth century the Shansabanis

Beginnings 113
adopted an avowedly Islamicate public presence, as they were increasingly

distinguishable from the many transhumant or/and nomadic groups of the
region. Implicit within this change of status was their ability to command
greater resources, some of which they dedicated to proclaiming their reli-
gious affiliation, albeit on a scale and in a context that was still far from the
cosmopolitan currents of the Persianate world.
Such a seemingly momentous and certainly consequential change
of lifeways and economies is by no means implausible. Among modern
nomadic populations of the Iranian world, it is important to note that
the variations between nomadism and sedentism can exist both between
affiliated or related groups, as well as within a single one, so that some
members are more sedentist than their transhumant and nomadic kinsmen
or others. Increasing association with settled or urban localities and sed-
entarization can take place for diametrically opposed reasons, namely,
the accumulation of surplus capital and purchase of land; or decline in
fortune and ultimate impoverishment, requiring employment by villagers
or other settled groups. Once undertaken, the process can be furthered
by the already more sedentist co-members of a group, who would provide
their transhumant or more nomadic comrades and kinsmen with ready
village or urban connections for labor opportunities, and sale or purchase
of supplementary or luxury goods.75 Such a process could have unfolded at
Qala‘-yi Zarmurgh, as the Shansabanis’ initiated their transition toward a
lifestyle less of seasonal transhumance, and more of an imperial formation
with seasonal “capitals.”
A question requiring focused discussion with regard to the early coa-
lescence of the Shansabanis is that of Islamization. Scholarly consensus
on Ghur’s Islamization has been that the region’s inhabitants were largely
hemmed in by a difficult, mountainous geography resisting both inte-
gration into wider commercial networks and the proselytism of Islam
(or any other widely dispersed socio-religious system), probably until the
eleventh-century Ghaznavid campaigns there.76 Prior to this period, “the
paganism of Ghur was of an indigenous variety and without outside con-
nections.”77 But probably by the early decades of the twelfth century,
the Shansabanis’ seemingly sudden patronage of Da Qadi’s minar – an
architectural form distinctly evoking Perso-Islamicate culture (if not
Islamic ritual) – requires a more nuanced explanation than the homog-
enizing and ultimately nondescript framework of “Islamization.” If the
Shansabanis required a transregional mercantile ethos and accompanying
religious culture for furthering their political and economic ambitions,
why not emulate the numerous Jewish mercantile communities through-
out Afghanistan and convert to Judaism, as did the ninth-century Khazars
north of the Caspian Sea? Islam was not the only choice.78
An important consideration must be brought to bear on the Shansabanis’
apparently instantaneous espousal of Islam: perhaps it was not altogether

114 Iran to India
instantaneous. Ghur itself surely did not remain immune to Islam and
particularly Perso-Islamicate cultural forms given the continuous presence
of both in neighboring Sistan and Zamindawar – presences beginning in
the eighth century and continuing uninterrupted into the twelfth: on the
heels of the Arab-Umayyad forays of the late seventh century, the entry
of the Arab-Abbasids there engendered volatile Kharijite reactions in the
eighth–ninth centuries, emerging particularly from Sistan.79 Thereafter
the Saljuqs and Ghaznavids continued the influx of Islam-affiliated ruling
ideologies and imperial presences into these regions.
During the first two decades of the eleventh century, the infiltration
of some form of Islamic religio-cultural presence in Ghur only increased.
According to C. E. Bosworth, the Ghaznavids’ own expediency in solidi-
fying their hold over the important commercial emporium and intellectual
center of Nishapur was instrumental in this process80: although Maªmud
Ghaznavi’s father Sabuktigin had been a supporter of the Karramiyya, this
populist sect’s appeal to and rousing of the city’s skilled labor groups – not
to mention their increasing antipathy toward Nishapur’s more orthodox
madhhabs – required their removal to the hinterlands. Bosworth proposed
that the sultan deputed the Karramiyya to the outlying areas of Ghur and
perhaps elsewhere, where their notorious proselytizing zeal would find
fertile ground among rural and transhumant/nomadic–pastoralist pop-
ulations (cf. also Chapter 4). In the end, however, Mahmud Ghaznavi’s
dislodging of the Karramiyya from Nishapur and their entry into Ghur
would have been only the most direct and documentable exposure to
Islam for the Ghuris, who were in all likelihood already exposed to Islam’s
ideologies and religio-cultural mores from at least the preceding century
or more.81
Within modern nomadic societies, anthropological studies have long
noted a “poverty of ritual activities” – and specifically their traces, that
is, ritual structures: nomadic life was structured around annual migra-
tions and temporary settlements rather than the observation of feasts,
fasts, and all-encompassing calendrical systems, especially if the latter were
unsynchronized with seasonal transhumance.82 In light of this overarching
nomadic tendency, it is plausible that, given the pre-imperial Shansabanis’
transhumance and/or nomadism, demonstrations of Islamic adherence
would be difficult to recuperate, as would many other social and cultural
aspects of their activities. Nevertheless, even if Islam co-existed with the
region’s “indigenous paganism,” it is not implausible that some Ghuris,
and particularly the Shansabanis’ pre-imperial ancestors, had already been
exposed to Islam for at least 200 years prior to the twelfth century, by
which time a predisposition toward this religious and cultural ethos could
have solidified.
By the beginning of the twelfth century and the construction of the
Da Qadi minar, the advantages of some degree of adherence to Islam –

Beginnings 115
or at least recognizable demonstrations of this adherence – would have

been increasingly evident to the ascendant Shansabanis. Although their
pre-imperial forebears had doubtless participated in the mercantile net-
works within Afghanistan, the truly transregional connections needed for
long-term commercial sustainability appear to have been plied largely by
the region’s Jewish communities, such as the one of Jam-Firuzkuh.83 But
in Afghanistan during the eleventh–twelfth centuries, empire-building –
essentially another mode of benefiting from commerce, with the added
advantage of being able to structure and direct it – lay in the hands
of confessionally Muslim elites such as the Saljuqs and Ghaznavids. In
order to become fitting competitors in an already crowded field of impe-
rial contenders, and to fill the vacuum left by the declining Saljuq and
Ghaznavid powers, appropriation and deployment of these fading prede-
cessors’ legibly Islamic – with a strong Persianate twist – imperial presences
afforded the best outlet for the Shansabanis’ own burgeoning political and
economic momentum.
The adoption of Islam, however, was not sufficient for effective king-
ship, particularly in light of the extremely variegated nature of both by the
beginning of the second millennium ce. Precisely because the Shansabanis’
transformation from autonomous or nominally ruled nomads and/or
practitioners of seasonal transhumance into expansionist rulers is not fully
retrievable from the surviving textual and/or material sources, another
look at the story related by Juzjani can provide at least the bases for further
insights.
Aside from Amir Banji, the other protagonist of Juzjani’s anecdote was
the cosmopolitan Jewish merchant from Ghur, who surely served as nar-
rative short-hand for the appreciable Jewish community of Jam-Firuzkuh.
As noted above, Jews were settled in the area from at least the early elev-
enth century and remained possibly through the last quarter of the twelfth
century,84 based on the dates in the Judeo-Persian inscribed tombstones
rediscovered at Kushkak, outside Jam. Taking into account the activities
of the other important Jewish settlements in Afghanistan, it seems that the
one at Jam-Firuzkuh would also have been prompted by mercantilism.85
But again, the location of Jam-Firuzkuh was less than propitious for
commerce: although the site lay at the confluence of the Jam Rud and
the Hari Rud, the rivers were volatile and seasonally difficult to cross.86
Moreover, the site was within a mountain fastness not easily accessed from
the well-plied routes connecting Herat, Balkh,87 and beyond toward the
northeast. Nevertheless, Jam-Firuzkuh’s Judeo-Persian funerary inscrip-
tions noted a variety of professions among the deceased, such as goldsmith
and banker. This variety in turn would indicate that the area might have
been equipped with at least some of the auxiliary “industries” and labor
specialties of banking and manufacture necessary for the full development
and exploitation of a mercantile economy. In the end, then, Juzjani’s

116 Iran to India
figure of the Jewish merchant likely personified a small but apparently

thriving Jewish community at Jam-Firuzkuh, established there by the first
half of the eleventh century and remaining active until the Mongol cam-
paigns of the 1220s.
Juzjani also highlighted in his story the friendly and mutually ben-
eficial interaction between Amir Banji and the Jewish merchant from
Ghur, probably evoking the tenor of relations between the pre-imperial
Shansabanis and Jam-Firuzkuh’s Jewish inhabitants. In studies of modern
transhumant and nomadic groups, large and small transactions have
been documented as de rigueur between them and farmers, villagers, and
townspeople along their seasonal migration routes. In fact, the necessary
regularity of these routes even enabled the mutual extension of credit.88
Thus, it is not implausible that, during the Shansabanis’ perennial tran-
shumance to and from their sard-sir “home,” they developed trading part-
nerships with the Jewish mercantile community already settled there. In
the end, Jam-Firuzkuh benefited the increasingly ambitious Shansabanis,
displacing Qala‘-yi Zarmurgh as a prominent sard-sir destination and
eventual summer “capital” by the mid-twelfth century. Jam-Firuzkuh’s
commercial–economic potential may have been uneven, but any access to
this potential required mediation, which was embodied in the locality’s
Jewish merchants and others with skills related to mercantilism.
Large nomadic confederacies from the Eurasian Steppes were undergo-
ing various types and degrees of Islamization already by the tenth century
ce, diversifying Islam itself as a political and religious ethos. Prominent
among these neophytes were the Saljuqs of greater Iran and eventually
Anatolia, who came to dominate the very centers of Islamic religio-political
authority embodied in the Abbasid khalifa at Baghdad. With the nearly
150-year Saljuq presence spanning from Baghdad via Isfahan to Marv
(mid-eleventh through twelfth centuries), nomadic lifeworlds combined
with the Abbasids’ Perso-Islamic imperial and administrative cultures
to create palpable distinctions between the western and eastern Islamic
lands.89 Although less of a direct threat to Baghdad itself, the Ghaznavids
represented another noteworthy group of empire-builders of the period,
viz. ancestrally Turkic military slaves whose fantastic territorial gains
earned them an enduring reputation over the centuries.90
Overlapping temporally and geographically with the Shansabanis,
the Saljuqs and Ghaznavids were not only their formidable competitors
in imperial expansion, they also served as aspirational models of preva-
lent courtly cultures (see Chapter One). Each of these three important
pre-Mongol political formations of the eastern Islamic lands accommo-
dated Perso-Islamic models of empire and imperial expansion on their
own terms, responding to their respective pre-imperial histories, the
demands of their political geographies, and the subjects constituting their
respective political formations. For the rising Shansabanis, then, similarly

Beginnings 117
contingent factors helped determine their adoption of some Perso-Islamic

mores – such as a fabricated pre-Islamic genealogy emerging from myth-
ical ancient Iran – alongside their continued adherence to indigenous
and non-Islamic traditions ensuing from the cultural economies of
transhumance/nomadism.
Notes
1. Asif 2016: 22 (see also below in main text).

2. For Qabacha see Juzjani vol. I, 1963: 418–421 (trans. vol. I, 1881: 531–544).
3. Juzjani related the anecdote of a courtier pointing out – rather audaciously
– that the sultan’s lack of sons meant his legacies had no heirs, to which the
sultan replied, “[While] to other sultans are [born] one or two sons, several
thousand are [born] to me, meaning the Turk slaves [bandagan], who will
inherit my kingdoms, and [they] will read the khutba in my name.” Cf. Juzjani
vol. I, 1963: 410–411 (trans. vol. I, 1881: 497); and Allegranzi 2019a, vol. I: 46.
See also Kumar (2007: 86ff.) for a description of the marital alliances among
the Turk ghulams presiding over the eastern territories (including Delhi),
which were instituted by Mu‘izz al-Din as the “patriarch” to mimic Islamic
marital practices – i.e., paternal cousin marriage – and establish inter-genera-
tional allegiances. Cf. also Chapter 6.
4. Asif 2016: 22.
5. Kufi (trans. Fredunbeg) 1900: esp. 95–98, 154–157; Siddiqui 2010: 28–29,
34–35; Asif 2016: 14, 86–92.
6. Habib (2017: 108), in his review of Asif’s book, adduced detailed evidence
maintaining Chach-nama as a translation of an earlier Arabic text; he sum-
marized as “a questionable thesis” the notion that it was Kufi’s original
composition.
7. Cf. esp. Asif 2016: 2–12 and passim.
8. The previously unknown geographical and historical origins of many commu-
nities in Afghanistan are gradually receiving attention. For example, the wide-
spread Ghorbat of Afghanistan – found in pockets spanning Sistan through
Qunduz, and Badghis to Jalalabad – were the subjects of a fascinating anthro-
pological study by Aparna Rao (1982). The author explored the possibility of
this and the Indian ja† origins of five other tribal confederacies. Although the
antiquity of the Ghorbat’s self-created ethnogenesis is unknown, it is note-
worthy that they also subscribed to descent from Shah-nama heroes (cf. Rao
1982: 41). One of the study’s important contributions is its documentation of
the negative connotations that ja† origins have throughout Afghanistan, prin-
cipally due to their nomadism (rather than their supposedly foreign origins).
Indeed, Bosworth (2011: 17) highlighted the contradictory descriptions of
Inner Asian Oghüz confederations in Arabic geographies, whose authors
characterized them as “purely steppe dwellers in felt tents, with no towns,”
even though substantial settlements of Turks and (Muslim) Iranians were not
uncommon in the same regions.
9. Cf. Amin 2002: 30; and Introduction.

118 Iran to India
10. See esp. Chapter 1, for a discussion of the Saljuqs’ historical adoption of Islam
during the tenth century ce and their later confessionalism.
11. In addition to the focus here on the twelfth century, Anooshahr (2018b: 2–3)
outlines the process for the fifteenth- and sixteenth-century emergence of
the global Islamic empires, but applicable here also: “… Persian historical
narratives reified and attempted to construct stable categories such as ‘kings’
[and] ‘dynasties’ … out of chaotic military–political events … invent[ing]
identities for the individuals involved (no longer as a band of armed men but
as ‘founders’ or ‘warrior-kings’).”
12. Cf. esp. recent studies by Cribb 2020; Cribb 2020 (forthcoming); also O’Neal
2016; O’Neal 2020. See also Chapter 1.
13. See esp. Ball (2020), who convincingly argued that the various zones encom-
passed within modern Afghanistan are unique for their array of “vigorous
pre-Islamic art and architecture,” attributable to the incursions of various
Hephthalite groups during the fourth century onward and the subsequent
expansion of the Western Turk empire (late sixth century). Cf. also infra and
Chapter 3.
14. Among the pioneering documentation of Ghur’s architectural remains
I include the photographs of Josephine Powell, available at: http://hcl.
harvard.edu/libraries/finearts/collections/photographers_archives.cfm. While
Powell’s photographs were complemented by Herberg’s documentation and
preliminary analyses (1978, 1982), Powell’s work continued to be useful for
Ball’s (2002) study of the region. See also Introduction and infra in main text.
15. Thomas 2018: 249 and passim.
16. While there is surely more information to be gathered, to date O’Neal’s
(2017) meticulous mapping of the Mongol campaigns in Afghanistan does
much to identify the zones likely experiencing the greatest impact from their
incursions. It is probable that there were at least three separate though simul-
taneous campaigns and nodes where more than one raiding party overlapped,
specifically (from east to west) at Parwan, Ghazna, Ahangaran, Jam-Firuzkuh,
Marv al-Rudh, and Herat. At least one raid probably reached Tiginabad/
Old Qandahar. Curiously, among the three raids converging at Ahangaran,
there was only one documentable offshoot southward into the heart of Ghur.
This raiding party appears to have reached the Rud-i Ghur’s headwaters in
the Yakhan Valley, closely following the Ghuris’ (including the Shansabanis’
ancestors) north–south migration routes. While Juzjani’s direct experience
of the Mongol campaigns in Afghanistan renders his Tabaqat among the
most important sources on the period, his patronage by Delhi’s Nasir al-Din
Mahmud (r. ah 644–664/1246–1266 ce) surely also played a role in the
author’s interpretations. Jackson (2017: 19) observed that Juzjani’s exagger-
ation of Mongol destruction “enabled [him] to portray the Delhi Sultanate
as the sole surviving bastion of Islam,” with the book taking on “at times an
apocalyptic tone.” Nizami (1983: 90–92), however, thought Juzjani’s first-
hand account of the Mongols was valuable, in particular his balanced portrayal
of Chingiz Khan. Cf. also Auer 2012: 19; Auer 2018: 101–102. But rather than
a Mongol monopoly on destruction, several centuries later other actors also
continued the erasure of Ghur’s historical traces: internecine turmoil among

Beginnings 119
the Durranis at Herat in the early nineteenth century brought destructive

repercussions to Ghur. Cf. Ferrier (trans. Jesse) 1976: 248ff.; and note below.
17. Bamiyan has always been on one of the principal routes joining the northern
and southern regions of the Hindu Kush, and also served as an administrative
center west of the mountains during both the pre-Islamic and Islamic periods
(A. A. Kohzad 1951a: 13; Klimburg-Salter 1989: 21; Klimburg-Salter 2008:
132; and Chapter 3). Its internal significance as well as being a major entrepôt
on the India–Central Asia artery off the trans-Asiatic silk routes rendered it
desirable for all local powers (cf. Klimberg-Salter 1989: 24), including the
early Shansabanis and their collateral lineage at Bamiyan (see below in main
text). Despite the great Mongol massacre at Bamiyan, however, it was not
among the nodes of convergence of the Mongol campaigns. Cf. O’Neal 2017;
Jackson 2017: 80, 158.
18. See also Thomas 2018: 39–44. The Ghuris, rather than having an endemic
antipathy to outsiders, were rightly affected by the far-reaching historical
ripples of the Great Game. Ferrier (trans. Jesse, 1976: 248ff.) noted that the
Ghuris he encountered were extremely suspicious of outsiders, Afghan and
foreigner alike, given the recent and continuing devastations unleashed on
the region from Herat: in the 1830s the city had been besieged by Qajar forces
under Russian guidance, while the western khanate of Herat was nominally
ruled by Kamran, son of Shah Mahmud (r. at Kabul 1800–1803, 1809–1818).
But his vazir Yar Muhammad had effectively usurped rulership from the
legitimate Durrani descendant and ordered the razing of all architectural
remains in the vicinity of Ghur’s famous citadel of Khissar (see below in
main text), out of fear that rebels and other enemies could quarter there and
eventually launch attacks against Herat from the east. Cf. Dupree [1997] 2012:
343–371.
19. A. A. Kohzad 1951a: 5.
20. Cited by Thomas 2018: 52. Juzjani vol. I, 1963: 328 (trans. vol. I, 1881: 318) of
course first described “the natural impregnability of the strong mountains
which are in Ghur.”
21. In his tours of Ghur, Kohzad (1954: 64ff.) also noted that the region’s
southern areas boasted more abundant villages than its northern ones –
understandable given that the Sistan plains were the over-wintering grounds
for many nomadic populations. This is surely a long-standing socio-economic
pattern, pertaining since at least the twelfth century when the Shansabanis
also migrated southward for the winters. See further infra in main text.
22. See A. A. Kohzad (1953: 59–62); O’Neal 2016a. Certainly, in the tour reports
of A. A. Kohzad (esp. 1953: 56) an intimate connection between historical
Ghur and Gharjistan to its northwest seems to emerge, underscored by these
regions’ shared semi-sedentist populations of the Tajik Firozkohi summering
there (also Jawan Shir Rasikh, pers. comm., December 2017). Indeed, modern
Abdali Pashtun nomads have also migrated across the provincial boundaries
between Ghur and Gharjistan (cf. Glatzer 1983: 215ff.). This strong inter-
regional connection could have spanned several centuries, as hinted by the
relatively early Shansabani presence in Gharjistan of the 1160s (noted supra
in main text) and their architectural patronage of the monumental Shah-i

120 Iran to India
Mashhad madrasa (1165–1176 ce). These observations are meant to underscore

what may be obvious, viz. that the historical frontiers of Ghur were porous
and even seasonal, requiring an expansive and flexible understanding of it and
its contiguous regions.
23. A. A. Kohzad 1954a: 23. See also O’Neal 2016a. As of the nineteenth century,
historical Ghur could be conceived as encompassing the areas of transhu-
mance and/or nomadic pastoralism of the Firozkohi and Taimani popula-
tions (see infra in main text and notes): the Firozkohis’ ambit extended from
Bala Murghab (modern Badghis Province) to the Hari Rud’s northern banks;
and the Taimanis’ from the Hari Rud as far as Girishk, near Lashkari Bazar.
Cf. esp. Maitland 1888: 605–608.
24. While A. A. Kohzad’s (cf. esp. 1951a, 1954a, 1954b) extensive reports on the
settlement of Ahangaran and the larger region are extremely valuable, I am
deeply grateful to Jawan Shir Rasikh (pers. comm., December 2017) for
sharing not only his photographs of the site and its vicinity, but also his
enlightening observations. See also Ball 2019: No. 15.
25. See esp. A. A. Kohzad 1954b: 13: though the ensuing description is of popula-
tions farther south in the Yakhan Valley (discussed below), the author made
the useful distinction between “migrant” (i.e., transhumant) and “nomadic”
populations: the nomads traditionally traveled much farther in seeking sea-
sonally hospitable climates, while their “migrant”/transhumant counterparts
moved only from built dwellings to collapsible ones such as yurts, largely
within the same area (e.g. Figure 2.12). Numerous parallels are also forthcom-
ing from Turkic nomadic populations, such as the Khazars (tenth century),
and the contemporaneous Oghüz (Golden 2013: 50–51, 54).
26. A. A. Kohzad 1953: 57ff.; also Barfield 1981: 97ff. for added data from north-
eastern pastoral nomads, who “are tied far more closely to the market economy
than are many … farmers.”
27. Gascoigne’s (2010: 145) analysis of ceramic finds from Jam-Firuzkuh left open
the possibility that, based on comparative material from Marv (Turkmenistan),
some unglazed coarse wares from the site might be datable even prior to the
eleventh century, thus before the earliest dates on Jam-Firuzkuh’s Judeo-
Persian tombstones. Cf. also infra in main text and notes.
28. Cf. Juzjani vol. I, 1963: 364 (trans. vol. I, 1881: 385–386); Thomas 2018: 149.
Admittedly, the notion that the Shansabanis’ summer rather than winter
settlement evolved into a “capital”-like locale creates a contrast to the sed-
entizing tendencies of most other transhumant and nomadic pastoralist
groups, whose winter encampments or settlements usually constituted their
initial foci of consolidated political power and the formation of ruling elites.
Cf. Golden 2013: 22, 30, 59.
29. This circuit of movement was possibly the residue of an il-rah, a collective
term referring to the prescribed routes of seasonal transhumance traceable to
the Shansabanis’ more transhumant lifestyle prior to the mid-twelfth century.
Taken from the fieldwork of Barth (1961: 4–5) among the Basseri confederacy
of tribes in southern Iran, the term il-rah encapsulates the affective concept of
“tribal road”: The il-rah in fact varied among the major nomadic groupings
of the region, each having their own traditional route of seasonal migration.

Beginnings 121
Some of the Basseri tribesmen who had gradually abandoned nomadism for
greater sedentism – mostly due to land ownership and agriculture – nonetheless
exhibited “a continuing emotional interest in and identification with nomad
life and ways” (Barth 1961: 106). Cf. also Barfield 1993: 96–97. In the case of
the imperial Shansabanis, the il-rah could have been more a convenience and
haptic trace of their pre-imperial ancestors, rather than a subsistence necessity,
perhaps even compelled by many nomadic societies’ ambivalence if not out-
right disdain toward sedentism (cf. Golden 2013: 22, 51).
30. Cf. Thomas 2018: 37. Tellingly, Thomas’s monograph (2018: 1–2 and notes,
146–150, also 315–319) confronts Jam-Firuzkuh’s lingering ambiguity in its
opening pages and closes with plausible explanations of it. I am grateful to
Warwick Ball (pers. comm., April 2020) for his observations on the two sites’
suitability as the summer “capital.”
31. Cf. Thomas 2018: 19, 40, esp. 149, for hypotheses regarding the Shansabanis’
garm-sir destination. Recently, W. Ball (pers. comm., April 2020) proposed
a few other sites as possible over-wintering “capitals” for the Shansabanis,
including Qala‘-yi Qaisar (Ball 2019: No. 875). Among the most promising of
these could be Shahr-i Kuhna, about 100 km south of Taiwara: not only does
it have the discernible arg-shahristan (round citadel–quadrangular settlement)
plan – indicating pre-Islamic origins – it is also in the vicinity of Ziyarat-i
Imam, a probable Ghaznavid-period shrine (ziyarat) of the late eleventh
century. See Thomas 2018: 253; Ball 2019: Nos. 1047, 1130, 1267.
32. The above-mentioned documentary tours of Afghanistan were all principally
carried out during the first three quarters of the twentieth century, as access
to the country as a whole became more limited after the Soviet invasion of
1979. Wannell nonetheless pursued an impressive itinerary in 1989–1990 (cf.
Wannell 2002). Constituting the bulk of documentation still available on
Ghur and many other parts of the country, these archives have been inval-
uable for me given the inaccessibility of large parts of Afghanistan during
my 2011 fieldwork. Cf. Thomas (2018: 113ff.) for a thorough summary of the
available data and interpretations of southern Ghur’s built remains.
33. Herberg 1982: 80–81.
34. Bosworth 1961; Edwards 1991: 90.
35. Juzjani vol. I, 1963: 332 (trans. vol. I, 1881: 331–332); see also O’Neal 2013:
40 and passim; O’Neal 2016a. An exception would be Herberg’s (1982: 83)
dating: based on what he saw as the stylistic and technical consistency of dec-
orative motifs on the surviving structures, he proposed that they were “part
of a chronologically restricted building programme, probably that of Baha
al-Din,” no doubt also relying on Juzjani’s report of this ruler’s directions to
building four fortresses on the borders of “Ghur, Garmsir, Gharjistan, and
the mountain tract of Hirat.” This dating would place southern Ghur’s forti-
fications in the mid-twelfth century, just after the ruler’s “completion of …
edifices and the royal palaces” of Firuzkuh. However, Herberg provided no
further evidence for his assertions in the form of comparable examples, or any
further reasons for a seemingly instantaneous fortification of the landscape in
this period.
36. Ball 2002: 41–44; see also Edwards 2015: 98–99. Although focused around

122 Iran to India
Awbeh and Chisht-i Sharif and elsewhere in the eastern reaches of Herat
Province, Franke and Urban’s rigorous surveys in 2004–2006 are still useful.
This work brought to light seventy-three previously undocumented struc-
tures, more than half of which lie in Chisht district. Nearer to Herat “large
sacral buildings” (Franke and Urban 2006: 6) – e.g., the Gunbad-i Shohada
and other ziyarats, shrines, and mosques – are numerous, but they decrease
steadily farther east: the east–west corridor along the Hari Rud, south of
Chisht, is peppered with what appear to be fortifications and/or surveil-
lance structures, including single towers (e.g., Kushk, Burj-i Qaria-yi Dehran,
Dara-yi Takht) and small complexes (Qala‘-yi Sarkari, Rabat-i Chawni),
all in all not unlike the architectural landscape of southern Ghur (cf. also
Chapter 4, below, and infra in main text). My own brief survey of areas
west of Herat toward Ghuriyan (2011) also revealed single towers and small
complexes. Altogether, these new findings demonstrate that areas other than
central–southern Ghur also evinced clusters of similar structures, though
the dating of all examples is still uncertain. See also Thomas 2018: 120–124.
But given the prominence of nomadic populations throughout Afghanistan’s
history (cf., e.g., Dupree [1997] 2012: 57–65), the remains possibly evoke
shared historical patterns where earlier fortification- or tower-strewn land-
scapes remained in enduring use by various populations over time (see also
Chapter 3).
37. Intriguingly, Juzjani (vol. I, 1963: 328 (trans. vol. I, 1881: 317–318) provided
a subtle hint that southern Ghur’s terrain was well-fortified by the ninth
century: evidently when the Saffarids emerged from Sistan onto a larger stage,
for our purposes reaching the region of al-Rukhkhaj, “the tribes of the Ghuris
fortified themselves on the summits of the rocks, and remained in safety; but
they used to be at constant enmity with each other … keeping up a war from
kushk to kushk …” Juzjani’s only utilitarian regard of architectural remains,
and complete disregard of their chronology – typical of most textual sources –
makes it difficult to know to which specific kushks he referred, or their dates.
Cf. Chapter 3, for a discussion of the term kushk, and more generally the
pre-modern receptions of architecture in the Persianate-Islamic world. As
a historiographical reflection, Juzjani’s usage of material culture primarily
as backdrop throws into even higher relief the present work’s focus precisely
on these elided historical indices, laying bare the very different methods of
recapturing the past in modern times.
38. Although the “dwelling tower” was identified as characteristic of Sistan
(Fischer 1970: 485), it serves to describe many of Ghur’s individual towers,
which also likely functioned as part-time residences (see main text). Overall
the survey and documentation of Sistan by K. Fischer and his colleagues
in the 1960s–1970s is relevant to our analysis of southern Ghur, not least
because the non-sedentist lifestyles prevailing among the region’s inhabitants
effectively blurred strict lines of delineation: akin to southern Ghur, Fischer
documented many ruined fortifications and individual towers throughout
Sistan, though pointedly noting that there were certain building types he con-
sidered regionally specific to it, along with the technique – especially in towers
– of combining the mud-brick structure with decoration in burnt or baked

Beginnings 123
brick. Cf. esp. Fischer 1971: 46, 50. Cf. also Ball and Fischer 2019: 487–505.
The Helmand–Sistan Project (HSP), directed by William Trousdale, under-
took further documentation in the region during the 1970s, which Mitchell
Allen is currently preparing for publication with W. Trousdale (M. Allen,
pers. comm., February 2020).
39. See esp. Herberg 1982: 75, 78; Thomas 2018: 114–116.
40. Cf. Herberg 1982: 73; Ball 2002.
41. See esp. Herberg (1982: 71) for a typological juxtaposition of the various
rectangular and cruciform plans of the structures he documented; this is
supplemented by Ball’s (2002: 29, 39) additional drawings of more complex
plans of single towers and small structures.
42. Cf. the analyses of Danestama (Le Berre 1970), a Buddhist site possibly repur-
posed during twelfth century (Shansabani?); Darra-i Killigan (Lee 2006);
and especially the pre-Islamic fort of Shahr-i Zuhak, characterized in broad
architectural terms as “fit[ting] into the sizeable group of pre-Mongol for-
tresses of central Afghanistan … for example, Chehel Burj and Qaisar” (Baker
and Allchin 1991: esp. 91 – these authors also had to rely on photographic
documentation of Ghur’s remains for comparative analysis). These complexes
are discussed in Chapter 3. The extensive ramparts of Balkh (ancient Bactres)
took full advantage of the site’s rocky promontory as the base on top of which
rammed-earth walls were erected (cf. Dagens, Le Berre and Schlumberger
1964). All of these sites’ distance from Ghur – located as they were either
within the ambit of the Bamiyan Valley and its offshoots, or far north in the
plains of Balkh – indicate that there was a recognizable pan-regional architec-
tural culture throughout most parts of Afghanistan.
43. I have relied on Szabo and Barfield’s (1991: 48, 68) documentation of the areas
of activity of the modern Taimanis and Firozkohis – both part of the Chahar
Aymaq confederacy – indicating their possible informativeness for historical
populations there. Additionally, P. A. Andrews’ detailed study of tentage has
posited historical data worthy of consideration, viz. the possible impacts of
Mongol and later Timurid tent traditions on those of the Chahar Aymaq (cf.
esp. King, vol. II, 1999: 155, 175, 363, 472, 682, figs. 6–8). Finally, the Afghan
Boundary Commission’s reports are extremely useful, providing a view on
the populations at least one century in the past: e.g., on the Firozkohis and
Taimanis as part of the Chahar Aymaq confederacy, see Maitland (1891:
107–141); Sing (1891: 161–204).
44. As intimated above, Szabo and Barfield’s (1991) work on indigenous domestic
architecture in Afghanistan has served as an invaluable resource here. See esp.
Szabo and Barfield 1991: 29–31, 49–50, 59–61, 69–71. According to Captain
R. E. Peacocke’s (1887: 148) observations during the 1880s on behalf of the
Afghan Boundary Commission, the Firozkohis considered their northern
boundary to be at Piwar near Bala Murghab (modern Badghis Province). The
Firozkohis’ circuit of seasonal transhumance in the late nineteenth century
was 130 miles east–west, and 45–80 miles in breadth, according to Maitland
(1891: 112).
45. Cf. Barth (1961: 101–102) for a discussion of the comparable limitations of the
nomadic–pastoralist Basseri of southern Iran.

124 Iran to India
46. It is worth noting that mid-twentieth-century archaeological surveys of

Ahangaran and its vicinity yielded unglazed painted ceramics initially dated
to “Kushano-Sasanian” occupation, which has usually meant c. 500 ce. Cf.
Baker and Allchin 1991: 107–109 and fig. 4.29 (reproduced from Leshnik
1968); Ball 2019: No. 15; Ball, Bordeaux et al. 2019: 391. This would have
impacted the significance of the site for our purposes, rendering it a pre-
existing and thus inherited site for the pre-imperial Shansabanis. However,
Franke’s (2016d: 235–236) recent analysis of comparable unglazed painted
pottery in the Herat Museum has resulted in assigning that entire collection
to the tenth–thirteenth centuries, concurrent with the glazed and other types.
The Herat examples that most closely parallel the Ahangaran finds include
Cat. Nos. PP120–PP127.
47. Cf. A. A. Kohzad 1954a: 29, 31; Wannell 2002: 236–237.
48. Herberg 1982: 80; (cf. also Thomas 2018: 116).
49. Both sites are among the many significant but scantily documented locales
throughout central and southern Ghur. Qala‘-yi Chahar Baradar was also
known as Pa’in Mazar, and Muna Ala as Mullah Ala and Mala Alau: cf.
esp. Ball 2002: 33–35; Ball 2008: 151–152 for a general description of the Ana
Valley, with Muna Ala in its western reaches; also Ball 2019: No. 742. For
Qala‘-yi Chahar Baradar, see Ball 2019: No. 787, and map 20.
50. Juzjani vol. I, 1963: 328, 331 (trans. vol. I, 1881: 318, 331). Mentioned in M. N.
Kohzad 1959; and published with analysis by Thomas et al. 2014 (also listed
in Ball 2019: No. 974). Thomas et al. (2014: 139) proposed that the “lost”
minar of Qala‘-yi Zarmurgh (discussed infra in main text) was most compa-
rable “geographically, stylistically, and perhaps chronologically … [with] the
mud-brick [sic] decoration on the Hephthalite and early Islamic fortresses
of Ghur,” indirectly suggesting a pre-Islamic date for the latter. As detailed
below, however, the decorative programs of these two corpora of structures
actually ensue from discrete architectural cultures.
51. For the Daulatabad minar, see Sourdel-Thomine 1953: 122–129; Ball and
Fischer 2019: 484–487. For the Qasimabad minar, see Tate [1910] 1977: 22
and facing, 270 and facing; and esp. O’Kane (1984: 97), who provided a span
of 1125–1150 ce for the minar’s construction, based on stylistic and epigraphic
comparison with more securely dated Saljuq towers, including the one at
Daulatabad. The dissemination of minars across Afghanistan adds further
weight to Najimi’s (2015: n. 46; cf. also Chapter 1) observations regarding the
mobility of specialized artisans, and their availability for patronage/employ-
ment at varying locales. See also infra in main text.
52. Many Saljuq minars, the two Ghaznavid examples, and, of course, the later
minar of Jam-Firuzkuh (ah 570/1174–75 ce) combined baked brick and
stucco in their epigraphic programs: while the Kufic or decorated (floriated)
Kufic tended to be executed in brick, the letters seem at times to be placed
against highly ornate, floral-arabesque backgrounds fashioned of stucco. The
naskh and other more curvilinear calligraphic styles were also executed in the
more pliant stucco. See Sourdel-Thomine 1953; Pinder-Wilson 2001; cf. also
Appendix.
53. A locality close to our area of study would be the Zoroastrian pilgrimage site

Beginnings 125
and irbidistan (Zoroastrian priestly institution) near Kuh-i Khwaja, now in

modern Iran’s Sistan and Baluchistan Province. The construction method of
stone foundations supporting sun-dried brick walls, widespread throughout
pre-Islamic Afghanistan, was prevalent here as well (Ghanimati 2000: 144).
Given the difficulty of pinpointing dates within the span of the Sasanian
centuries – particularly in heavily trafficked landscapes such as the historically
well-traversed Sistan – it is unfortunate that the site’s sample collections
were not more thoroughly published (fieldwork was conducted separately
by A. Stein and E. Herzfeld in the early twentieth century; for the Herzfeld
reproductions see Kröger 1982: pls. 23/6 [painting], 103/1–2, 104/4 [stucco]).
Further sampling and 14C analysis in the 1990s resulted in the discernment
of two or three construction phases for the complex, ranging from the late
Parthian through late Sasanian periods (second–seventh centuries). Cf.
Ghanimati 2000; Canepa 2013. However, Kröger (2005) dated the site to the
“early Sasanian period” or c. third–fourth centuries ce. For the other sites,
see Schmidt 1937: 337–338; also Kimball 1937: 347–350, fig. 173, pls. LXXIV–
LXXV; Kröger 1982: pls. 88, 1–7 and 89, 1–7; Kröger (2005); Boucharlat 2019:
362–363.
54. Even with important discoveries during the intervening decades, Kröger’s
1982 monograph still serves as the main reference for Sasanian stucco finds
from across the empire, representing a transregional study of the medium
throughout the Sasanian lands. See also Kröger 2005; Boucharlat 2019: 351.
55. It should be noted that Kröger (2005) posited stylistic and iconographic
echoes of late Sasanian stucco from sites such as Nizamabad and Tepe Mil in
that of early Islamic palace complexes (e.g., Khirbat al-Mafjar, late seventh–
early eighth centuries ce). Characterizing the process as one of “both a con-
tinuity and a change” initiated by the work of “northern Persian” artisans
working in Palestine, it forms a possible parallel to the processes unfolding in
the eastern reaches of the Sasanian realms during the seventh–tenth centuries
as noted here.
56. Cf. Bosworth 1994: 9ff.
57. Marquart, relying on observations gleaned from al-Biruni (cited in Bosworth
1961: 124), considered the usual root of RTBL in Arabic sources as erroneous,
arguing instead for “Zunbil” as a title (not a personal name) of a ruling
lineage with lingering Hephthalite or Western Turk origins. The repeated
occurrence of the term in Tarikh-i Sistan ((trans. Gold) 1976: esp. 74 n. 1,
170) hints at a localized nomenclature. But most scholars now favor the
RTBL root over Marquart’s reading, as the likely Arabicization of the Turkic
title iltäbär: see Bosworth 1968a: 34–36; Bosworth 2008: 97–99, 106–109;
Ball 1988: 134–135; Ball 1996: 399; Inaba 2010; Gardizi (trans. Bosworth) 2011:
128, n. 18; Rezakhani 2017: 165ff; N. Sims-Williams (cited by W. Ball, pers.
comm., April 2020); Ball 2020. Notwithstanding the etymological arguments
above, al-Biruni’s characterization of Zamindawar’s famed shrine of Zun/
Zhun as dedicated to a minor sun deity can be circumstantially confirmed.
Zamindawar lay on well-trodden commercial routes linking Panjab (and ulti-
mately the subcontinent) with the greater Iranian world. Multan in southern
Panjab was not only a significant commercial node but also home to the

126 Iran to India
famous Brahmanical temple dedicated to the solar deity Aditya, and directly
communicated with Zamindawar; Brahmanical trends thus easily traveled
westward and likely resulted in derivative cults. For Multan’s regional and
transregional overland connections, see Deloche [1980] 1993/94: 26–27 and
fig. II; Dani 1982; Rehman 1979: 66; Edwards 2006: 22; and esp. Chapter 6.
The mutually augmentative nexus between pilgrimage and trade has been
extensively studied, e.g., Ray (1996: 5 and passim).
58. Inaba 2013: 90; Rezakhani 2017: 140–145, 165–168; and Chapter 3. The com-
merce along these routes, moreover, invited raids by Ghuri tribesmen along
the southern reaches of their own annual migrations. Cf. also Bosworth 1961:
118; Mahmud 2009: 54; and Chapter 1.
59. Cf. Anon., Tarikh-i Sistan (trans. Gold) 1976: 74–75, 86–87, 101, 111, 114,
119, 167, 170–172, 216. Apparently, when central Umayyad demands became
oppressive, there were even periodic alliances between local Arab forces and
the Rutbil (see ibid., esp. 89–96, 163–164). See also Bosworth 1968a: 34–36,
121; Ball and Fischer 2019: 460; Ball 2020.
60. Michailidis 2015: 136. See also Ball 2020. Later patrons “could have indeed
understood [the motifs] to reflect an earlier tradition” (M. Canepa, pers.
comm., February 2018). Parallel to Michailidis’ characterization of the archi-
tectural patronage of the Bawandids (mid-seventh–fourteenth centuries), a
similar continuity with Sasanian architecture was discernible in ˝ahirid (821–
873) palatial constructions (cf. Finster 1994). Subsequently, the well-known
Buwayhid (932–1062) inscriptions at Persepolis demonstrated the importance
of pre-Islamic Iranian kingship especially for dynasties with local origins,
including the adoption of titles such as shahanshah. Cf. Blair 1992: 32–37.
See also Canepa’s (2015: 92–99) analysis of Sasanian dynastic sanctuaries and
the dissemination of “Iranian kingship as a global idiom of power.” Finally,
Bosworth (1968a: 23) confirmed the proclivity toward Sasanian imperial
practices specifically in Sistan, where “the local consciousness … had strong
feelings of solidarity with the ancient culture and traditions of Iran.” See also
ibid., 35 for the prestige and emulation of the Sasanian empire throughout
Central Asia during the centuries after its demise; and Bosworth 2008: 97ff.
It should also be mentioned that Sasanian coinage types continued to be
emulated in the empire’s eastern lands, possibly as late as the eighth century
(Rezakhani 2020).
61. Cf. Bosworth [1977] 1992: 68; and Introduction (notes).
62. In terms of architectural patronage, regional variations among the far-flung
Saljuq domains were discernible for a variety of reasons, perhaps especially
due to the increasing politico-economic fragmentation of the empire through
the iqta‘ system (Peacock 2016: 19–21). Nevertheless, the iqta‘ holders, some
becoming more independent of the Saljuq sultans over time, “also sought
to enhance their own prestige by creating regional courts modeled on the
Seljuk precedent, which acted as centers for artistic and cultural patronage”
(ibid., 15).
63. While the figure of the Jewish merchant from Firuzkuh has had multiple his-
torical resonances – discussed forthwith – recent re-examination of the name
“Banji” has also borne fruit as to its possible origins: analysis of pre-Islamic

Beginnings 127
coinage bearing mint names ranging from Zabul through Balkh has led
Rezakhani (2020) to propose that “Banji” might be an Arabicized corruption
of “Pangul,” one of the regional rulers named on the aforementioned coins.
As for the Jewish merchant, Fischel (1965: 149–150) also relied on the figure as
a point of departure for his analysis of the Judeo-Persian funerary inscriptions
from Jam-Firuzkuh’s Jewish cemetery, a small group of which were rediscov-
ered in the 1960s, with more subsequently coming to light. Approximately
seventy-four funerary markers – carrying a total of at least ninety-one Judeo-
Persian inscriptions – were found over the course of several decades at a short
distance from the famous minar of Jam-Firuzkuh, in the Jewish cemetery cut
into the mountainside at Kushkak. They commemorated personages of dif-
ferent occupations, with dates of death starting in the early eleventh century
and continuing through end of the twelfth. See also Habibi 1980: 42; Pinder-
Wilson 2001: n. 37; Lintz 2008; Lintz 2009; Hunter 2010; Thomas 2018: 214ff.
The tripartite mural inscription at Tang-i Azau, about 200 km east of Herat
and also in the vicinity of Jam-Firuzkuh, is datable to the fourteenth century.
Cf. Ball 2019: No. 1144. The implications of the existence of a Jewish com-
munity at Jam-Firuzkuh, prior to its designation as the Shansabanis’ summer
capital in the mid-twelfth century, are explored below.
64. Juzjani vol. I, 1963: 324–327 (trans. vol. I, 1881: 311–316); also Edwards 2015:
98.
65. Cf. Juzjani vol. I, 1963: 328, 331 (trans. vol. I, 1881: 318, 331); for a discussion of
the relevant passages from Juzjani, see also Thomas et al. 2014: 139–140.
66. According to the local reports gathered by Thomas et al. (2014: 136, 139–140),
the ruins of a mosque “over 100 years old” lie adjacent to the tower; its
salvageable wooden elements – mainly pillar bases – have been incorporated
in the new village mosque as a reminder of the no longer usable structure.
(Incidentally, the mosque’s dating to the nineteenth century is supported by
the lines of poetry by the Turkish poet Hafiz Burhan (1897–1943) inscribed on
the aforesaid wooden components.) The frequent association of minars with
mosques may hint at the replacement of an older structure in the nineteenth
century. Moreover, the minar and mosque ruins are by no means the only
probably historical remains in the vicinity: on an adjacent promontory lie
the foundation stones of a ruined fortified palace, locally known as qasr-i
dukhtar-i malik (“castle of the king’s daughter”).
67. Thomas et al. 2014: 138.
68. Cf. esp., e.g., Sourdel-Thomine 1960; Blair 1985.
69. Cf. Ball and Fischer 2019: 495–498; Ball 2020; and Volume II. The Sistan
towers lack any remaining epigraphic data and so provide no indication of
patron or date, but have been assigned stylistically to the eleventh (Pinder-
Wilson 2001: 173) or twelfth century (O’Kane 1984: 100); the earlier date
may be more credible based on the handling of the brick revetment, and the
comparability of the towers’ stellate plans to those of eleventh-century Saljuq
tomb towers (e.g., Gunbad-i Qabus, dated by inscription to ah 399/1009
ce). Thus, there was more than one formal precedent within Afghanistan
even for minars. As with the Qasimabad minar, the Nad-i Ali tower did not
figure in Fischer et al.’s documentation of Sistan’s archaeological remains in

128 Iran to India
the 1960s–1970s, indicating that it too had collapsed by this time. A print
of Tate’s photograph of the Nad-i Ali tower remnant (1910–1912: 202 and
facing) is also in the Royal Geographic Society’s collection (London). Cf. also
O’Kane 1984: pl. XIVc.
70. Sistan as a source for the skilled laborers required to build the minar at Da
Qadi can be supported by its proximity. However, according to the exten-
sive surveys and documentation by Fischer (and his larger team) throughout
Sistan, the region was distinguishable from many other parts of Afghanistan
in the local architecture’s frequent combination of mud and burnt brick, par-
ticularly in the decorative elements. Cf. Fischer 1971: 50; Fischer et al. 1974:
illus. 252–254; Tate [1910] 1977: 22, 270 and facing figures; O’Kane 1984: 90.
Indeed, the possibly early date and peculiar form of the region’s brick towers
– cf. above in notes – only reinforce Fischer’s observations regarding Sistan’s
architectural distinctiveness.
71. For Ghazna, see Pinder-Wilson 2000: 155. Tate ([1910] 1977: 271) estimated
the Qasimabad minar to be “one hundred feet [c. 30 meters] … at the
outside.” Hillenbrand (1994: 148) confirmed that Saljuq minars generally
reached heights of about 30 meters. Cf. also Bloom 1989: 170–172.
72. Thomas et al. 2014: 137.
73. For Ghazna as a comparative example, see Appendix, Afghanistan VI.
74. Thomas et al. 2014: 138.
75. N. Yoffee, pers. comm., November 2017; also Barfield 1993: 106–107. Barth
(1961: esp. 105–110) documented just such a process of “patchwork” seden-
tarization among the Basseri, i.e., affecting only parts of the self-identified
group, rather than en masse. Sedentarization among nomadic groups is a
known phenomenon (cf. also Golden 2013: 30), though the processes’ specifics
differed in each instance, depending at least in part on a region’s unique
circumstances, including economic bases and travel networks, both of which
the nomads helped to shape or shift over time depending on their activities.
These specifics and the economic factors prompting them, as well as the larger
consequences for nomadism in general, continue to be debated: see Barfield
1981: 165–170 for a succinct differentiation of studies on nomadism in Iran
(esp. Barth 1961), whose Basseri subjects have been considered the most com-
parable example here) and Afghanistan.
76. E.g., Mahmud 2009: 60.
77. Bosworth 1961:125.
78. Here, I take up Jackson’s (2017: 333 and passim) meaningful chapter sub-
heading “The Choice of Islam.” The author has provided a significant (if
brief) overview of the various contingent factors in religious conversion,
demonstrating that adopting Islam – or indeed any other transregional socio-
religious system – as the “state” religion was the result of many historical
and regional contingencies. In the case of the several Mongol khanates, for
example, shifting from adherence to and patronage of their small-scale ritual
practices to larger religious systems varied widely, with the eastern khanates
opting for Tantric Tibetan Buddhism precisely as a statement of contrast
from their mainstream Chinese Buddhist subjects. Thus, “it was not necessar-
ily the case … that the Chinggisid princes of Western Asia would succumb to

Beginnings 129
… the Muslim majority. Realpolitik could instead have dictated moving in a

completely different direction” (ibid., 334). In parallel fashion, Crossley (2019:
87–88) explored the array of religio-cultural choices and their consequences
among the Bulgars. For the Khazars and Judaism, cf. Amitai 2008: 283–284
and notes; Kaplony 2008: 320–321; and esp. Golden (2013: 53–56), who noted
that, parallel to the Shansabanis’ “origin myth,” the Judaicization of the
Khazar ruling elites – encompassing both the qaghan or sacral king, and the
subordinate governing king (Ar. malik) – had also begun during the reign of
Harun al-Rashid.
79. See esp. Bosworth 1994: 67–69, 79–80; Siddiqui 2006: 19–20.
80. Bosworth 1961: 128–129.
81. See also Gardizi (trans. Bosworth) 2011: 57; Bosworth 2008: 100; Bosworth
2012c. For the Karramiyya in Nishapur and Ghur, cf. Bulliet 1972: 12–13,
68ff., 71ff.; Zysow 2012; and Chapter 4. Regarding the Karramiyya’s pros-
elytism in particular, Amitai (2008: 283–284) noted that, while they were
“the best organized group of Sufi character … [with] khanqahs in Farghana,
Khuttal and Samarqand,” they actually were not very successful agents of
Islamic conversion among the Turkic peoples of Transoxiana during the
tenth–eleventh centuries. Much needed re-examination of the long-held
notion of Sufis as the major force in the Steppe Turks’ Islamization has led
scholars to propose instead that traveling Muslim merchants, a few select
political–military figures, and even mainstream but itinerant fuqaha‘ were
the more likely purveyors of exposure – with only eventual conversion – to
Islam among the Turkic peoples. See Flood 2005a: 272 and passim; cf. also
Chapter 4.
82. See esp. Barth 1961: 135–136; Barfield 1981: 76–77; Amiri 2020: 151–152.
Gellner (1983: 446) additionally advocated for “… a tribal Little Culture,
which does not meet the standards of an old Great Tradition, [and] should
not be denied for the past simply because it no longer satisfies the needs of a
more urbanized, centralized, literate society which has replaced it.”
83. During the first and second millennia ce in West through Central Asia (and
beyond), identifiable religious communities – such as Manicheists, Buddhists,
Nestorian Christians, and Muslims – had constituted major mercantile forces
within larger polities. Their commercial power lay largely in the ability to
partner with co-religionists, kinsmen, and others to create and utilize vast
transregional networks for the circulation of goods and capital. The “cor-
porate” integration of these mercantile networks made religious conversion
advantageous, as in the case of the Uighur conversions to Manicheism, and
the Bulgar conversions to Islam (see Jackson 2017: 333–334). It was perhaps
due to the absence of such identifiable modes of “incorporation” (i.e., conver-
sion) into “Hinduism” that the networks of Gujarati, Panjabi, and Marwari
merchants in Central Asia did not achieve the prominence and longevity of
other groups. Cf. esp. Dale 1994; Levi 2002.
84. While the presence of Jewish communities throughout Afghanistan – attested
at Bamiyan and Ghazna (cf., e.g., Haim 2012; Haim 2019) and elsewhere –
Jam-Firuzkuh’s tenuous geographic connectivity and “the dearth of clearly
stratified occupation” led D. Thomas (pers. comm., October 2018) to raise the

130 Iran to India
possibility that the tombstones might have been posthumous memorials to

venerated ancestors, rather than collectively indicating a sizeable Jewish settle-
ment in the area. Nevertheless, it should also be noted that the apparent elev-
enth-century terminus post quem of the Jam-Firuzkuh Jewish community is in
line with the evidence of Jewish communities at other important commercial
cities/emporia in Afghanistan, such as Bamiyan, whence a sizeable collection
of letters between Jewish family members have begun to provide invaluable
information regarding their mercantile interests, as well as religious, liturgical,
legal, and even poetic activities. Cf. http://web.nli.org.il/sites/NLI/English/
library/news/Pages/Afghan-Geniza.aspx. Documents within this collection
also furnish information about Jewish groups settled at yet other Afghan cities
such as Ghazna as of the eleventh century: see Haim 2012.
85. The anecdotal figure of the Ghuri Jewish merchant served as an important
historical clue for Fischel’s (1965) and others’ initial analyses of the Jam-
Firuzkuh tombstones. See also Moline 1973/74: 147–148. For other studies
of the inscriptions and the Jewish communities of Afghanistan, cf. inter alia,
Fischel 1965; Pinder-Wilson 2001: n. 37; Lintz 2008; Lintz 2009; Hunter
2010; Thomas 2018: 214ff.; and above in main text and notes.
86. See esp. Thomas 2018: 54, 174, 189. Indeed, the disastrous flood of the Hari
Rud, probably c. ah 595/1199 ce, which caused the destruction of the mosque
adjacent to the famous minar of Jam-Firuzkuh, was noted by Juzjani, vol. I
1963: 375 (trans. vol. I, 1881: 404).
87. Given the significance of Balkh (ancient Bactra) in northern Afghanistan as
“the emporium of India” at least since the ninth–tenth centuries, it is little
surprise that an established Jewish mercantile community resided within the
rabad – just outside the madina, though all still within the fortified city – as
indicated by the Bab al-Yahud. Similarly, the Bab al-Hinduwan signified an
Indian merchant community also in residence at Balkh. Cf. Bosworth 1988.
88. Cf. Barth 1961: 9–10, 97–100, 102; Barfield 1981: 28, 97ff.; Barfield 1993:
94–95, 99–100; Mahmud 2009: 251ff. See also Barfield (1981: 102; 1993: 98)
for the contentious – though still regular – relations between nomads and
farmers with respect to the damage done to crops by the former’s grazing
animals.
89. Although providing parallels to the Shansabanis’ nomadism and participation
in Perso-Islamic forms of kingship, clearly the Saljuqs did not go from ruled
to rulers, but rather “inundated the Iranian world en masse, in their tribal
and clan units” (Vàsàry 2015: 16). Moreover, Abbasid expansion into Iran and
beyond during the late eighth century had already brought “eastern” Sasano-
Iranian courtly practices into the Islamic world (see Finster 1994: 17–18). It
was Saljuq political and military control of the Abbasid caliph and his capital
at Baghdad, however, which created a more palpable east–west conduit, insti-
gating the arrival of intellectuals and others westward to Baghdad. But rather
than absorbing Baghdad into a fulcrum of power effectively shifting eastward
toward the Saljuqs’ preferred base at Isfahan, these migrations actually reified
the political, social, and cultural differences between Baghdad and the Saljuq
“east.” Cf. Durand-Guédy 2013b: 325–326, 332–334; Van Renterghem 2015;
also Peacock 2015: 134ff., 163–164, 168–169.

Beginnings 131
90. The reputation of the Ghaznavids appears to have been founded largely on
their India campaigns, which served multiple purposes such as gaining public
admiration and the support of the Abbasid khalifa at Baghdad, in turn further
spreading the dynasty’s renown. Cf. Bosworth 1966: 87ff. Recent analyses
of the textual sources reporting events during Ghaznavid ascendancy have
underscored the multiple motivations behind the India campaigns, empha-
sizing that facticity was less the authors’ goal than communicating the poly-
valent meanings of these campaigns in their own time. Cf. Anooshahr 2006;
Anooshahr 2018a.

CHAPTER 3
The Early Shansabanis:

Firuzkuh, Bamiyan, and Tiginabad/
Old Qandahar, c. 1140s–1170s
I n c. ah 539–540/1145–1146 ce, Qutb al-Din Muhammad (d. ah 543/1148

ce), ‘Izz al-Din’s second son (Genealogy, p. xvii), established his seat
at Firuzkuh – now generally accepted to be near the village of Jam, at the
confluence of the Hari Rud and Jam Rud (Figure 2.2).1 In virtual sim-
ultaneity, the Shansabanis expanded eastward beyond Ghur to the great
religious center and commercial entrepôt of Bamiyan, where Fakhr al-Din
Mas‘ud (r. ah 540–558/1145–1163 ce), ‘Izz al-Din’s eldest son, presided
over an agnatic Shansabani lineage that ruled the region through the early
thirteenth century. In the late 1150s, the infamous ‘Ala’ al-Din Husain
Jahan-suz (r. ah 544–556/1149–1161 ce; further discussed below) departed
from Firuzkuh southward and wrested Qandahar – quite possibly the
medieval Tiginabad (see later in this chapter) – from his arch enemies the
Yamini-Ghaznavids.2 It was from here that, little more than a decade later
in ah 568/1173 ce, the brothers Shams al-Din Muhammad and Shihab
al-Din Muhammad – two of ‘Izz al-Din’s grandsons, and nephews of
the notorious ‘Ala’ al-Din Jahan-suz – would set out to eradicate the
Oghüz bands occupying Ghazna. Upon achieving this victory, Shihab
al-Din was crowned Sultan Mu‘izz al-Din there, thus creating a third
Shansabani lineage at the once grand, erstwhile Yamini capital (cf. esp.
Chapter 5).
The present chapter’s focus on the late 1140s through early 1170s ce –
prior to the definitive Shansabani conquest of Ghazna ah 569/1173–1174
ce – encompasses the initial phase of Jam-Firuzkuh and the early Shansabani
campaigns to Bamiyan and Tiginabad/Old Qandahar. Against the back-
ground of this multi-directional expansion, the Shansabanis’ architectural
patronage in the successively acquired power bases is extremely varied,
defying any expectations of a recognizably unified and singular imperial
presence. In fact, this variety seems to belie centralized imperial expansion
(cf. also Chapter 1, above).
Archaeological investigations at the site of Jam-Firuzkuh brought forth
virtually no evidence of palatial architecture from the time of Shansabani
occupation (1140s–c. 1215), despite Juzjani’s mentions of large palaces and
forts there. Instead, the early Firuzkuh Shansabanis seem to have con-
centrated on defensive and religious–civic structures (Figures 3.1–3.6; see

The Early Shansabanis 133
below). As mentioned in Chapter 2, and argued more thoroughly in this

chapter, more than a conventional capital city Firuzkuh was likely a royal
sard-sir (summer) encampment – internally and externally well-defended,
to be sure (Figures 3.2 and 3.3; see below): its imposing minar (Figures
3.5 and 3.6) was one of probably very few monuments per se marking the
locality. At the same time, the Bamiyan region (Figures 3.7–3.13) boasts
perhaps the broadest array of structures evincing Shansabani-period inter-
ventions. But rather than a singular mode of building, Bamiyan and its
contiguous valleys encompass an appreciable architectural variety, ranging
from reuse and repurposing of pre-existing complexes (Figures 3.9–3.13), to
large-scale new constructions (Figures 3.24–3.27). Old Qandahar (Figures
3.28 and 3.29) rounds out the Shansabanis’ arc of initial expansion and is
even more enigmatic than the other sites: here only funerary mounds can
be associated with Shansabani presence.
Although the Chisht, Gharjistan, and Ghazna-Bust region are dis-
cussed more thoroughly in Chapters 4 and 5, they merit brief mention
here: the nomadic Oghüz had occupied Ghazna and its vicinity for over a
decade by the time the Firuzkuh forces effected their momentous victory
there in ah 568/1173–1174 ce (see below). But even after this long period
of neglect at the old Yamini capital, and likely active deterioration during
the nomadic bands’ presence, Ghazna and Bust-Lashkari Bazar still pre-
sented usable architectural fabrics, which apparently minimized the need
for new Shansabani-patronized constructions.3 It is only in the western
and northern directions from Jam-Firuzkuh at Chisht and Gharjistan that
the Shansabani presence left behind newly commissioned monumental
architecture.
Doubtless, some of the varying methods of making an imperial mark
were inflected by the Shansabanis’ pre-imperial transhumant and/or
nomadic beginnings, which continued in modified form into later years
(see Chapter 2). These lingering mores surely combined with the exigen-
cies of the occupied territories’ own pre-existing landscapes, resulting in a
noticeable variety among the early Shansabani annexations. The analyses
in Chapters 2–5 of the surviving built remains, then, are essential not only
for obtaining a complete picture of the Shansabanis’ early imperial expan-
sion; they are equally so for a thorough understanding of the disparate
cultural and historical geographies throughout Afghanistan during the
twelfth century. Bamiyan, Tiginabad/Old Qandahar, and the outward
routes from Jam-Firuzkuh to Chisht, Gharjistan, and Balkh – all inte-
grated to varying degrees within the age-old Khurasani commercial net-
works radiating in multiple directions from Herat – had each undergone
very different developments from the early centuries of the Common
Era (and before). These inherited circumstances in turn impacted each of
the respective areas’ distinct later significances for local and transregional
imperial interests well into the second millennium ce. It is worthy of

134 Iran to India
remark (even in hindsight) that the co-existing differences among these

Shansabani presences outside Ghur seem to serve as harbingers presaging
the later, larger-scale and yet distinctly region-specific modes of “con-
quest” along the banks of the Indus and beyond (cf. Chapter 6, and
Volume II).
As discussed in Chapter 1, the Shansabanis’ quickly increasing sphere of
influence was not necessarily the outcome of a strategy emanating from a
“center.” Rather, it seems frequently to have been the result of ad hoc and
piecemeal campaigns from the newly invested appanages. Their architec-
tural patronage would buoy this understanding. But, recreating the precise
process of expansion is made all the more difficult by various aspects of
the textual sources closest to the events. The prime among these sources
would of course be Juzjani’s Tabaqat of the mid-thirteenth century, along
with surviving near-contemporaneous works belonging to various other
genres.4
One challenging aspect of Juzjani’s Tabaqat (as well as other texts)
may seem merely mechanical, but it is far from immaterial. In addition
to the still cautious identification of Firuzkuh,5 the majority of places
Juzjani named as the Shansabani brothers’ appanages are as yet only con-
jecturally located. For example, the precise whereabouts and extent of
the territory of Warshada, the appanage of ‘Izz al-Din’s second son Qutb
al-Din Muhammad – the eventual “founder” of Firuzkuh – are unknown.6
Although, Qutb al-Din’s subsequent search “in [various] directions for
a place to build a well-fortified stronghold and a wondrous city as a
capital” could indicate that Warshada was in the vicinity of Jam-Firuzkuh.
Ultimately, Qutb al-Din’s selection of this locale as his seat attested to his
acuity and foresight, given the more stable, commerce-based prosperity of
the settlement and its vicinity.7
Aside from Warshada’s hypothetical location, only Kashi and
Wajiristan can be identified. Kashi was the appanage of Fakhr al-Din
Mas‘ud, ‘Izz al-Din’s eldest son, and can be surmised as a part of or near
modern Chaghcharan (capital of present-day Ghur Province). Wajiristan,
the appanage of ‘Ala’ al-Din Husain Jahan-suz (the “World Incendiary”;
see below), was apparently located directly west of Ghazna. But the appa-
nages of ‘Izz al-Din’s other sons – viz. Madin, given to Nasir al-Din
Muhammad; Sanga to Baha’ al-Din (d. ah 544/1149 ce); and even the
Istiya of Saif al-Din himself – are only approximately locatable, if at all.8
Early Firuzkuh
By the time of ‘Ala’ al-Din Husain Jahan-suz’s vengeful and supposedly

spectacular destruction of Ghazna and Bust-Lashkari Bazar in ah 544/1150
ce, his older half-brother Qutb al-Din had already initiated a Shansabani

hold over Firuzkuh. As detailed previously, due to Qutb al-Din’s pre-

mature death, Firuzkuh was left to his agnatic successors and ultimately
inherited by ‘Ala’ al-Din himself.
In contrast to the abundant information in Juzjani’s Tabaqat regarding
military confrontations and political alliances, the work offers little on
Firuzkuh as a settlement or its architecture, such as the population or
extent and general location of its buildings. In the seventeenth tabaqa
on the Shansabani sultans and maliks of Ghur, the author mentioned
Firuzkuh only infrequently, referring to it as “a fort and a city” without
further detail. He does note that after Qutb al-Din’s decampment to
Ghazna, Baha’ al-Din “edified the city and completed the royal buildings
and the palace (qasr).”9
Only an intimation of the ruler’s residence and central area of Firuzkuh
can be gleaned from Tabaqat, and that also from multiple parts spanning
several sultans’ reigns over approximately sixty years. The majority of the
information – meager as it is – comes from the later, imperial phase of the
site (cf. also Chapter 4): in the entry on the penultimate Shansabani sultan
Ghiyath al-Din Mahmud (r. ah 602–606/1206–1210 ce), Juzjani reported
that, during the time of this ruler’s more renowned father Sultan Ghiyath
al-Din Muhammad (r. ah 558–599/1163–1203 ce), the qasr known as Baz
Kushk-i Sultan (“the hilltop [called] ‘the Sultan’s Pavilion/Palace’”) was
located in the middle of the city. Furthermore,
… the qasr is a building whose equal is not found in any other kingdom or
in any capital … in height and area, and with columns (arkan), belvederes
(manzar-ha), galleries (riwaqat), and towers/crenellations (sharfat) no engineer
has created [before]. On top of the qasr are five golden turrets (kungura), each
one three gaz and a little over in height, and two gaz in circumference; and also
two golden huma, each the size of a large camel. After the conquest of Ajmer,
Sultan-i Ghazi Mu‘izz al-Din Muhammad had sent those golden pinnacles
and huma to Sultan Ghiyath al-Din’s capital as a token of his servitude, along
with much else of rarity and value, such as a golden ring with a golden chain
and two (golden) orbs (kharbuza, lit. “watermelon”) that were five gaz by five
gaz, and two golden kettle-drums (kou) that were brought by cart … Sultan
Ghiyath al-Din ordered the ring and chain, and those orbs (kharbuza) to be
suspended in the monumental entrance portal (pishtaq) of the masjid-i jami‘
of Firuzkuh …10
From this brief passage, it is apparent that Firuzkuh’s masjid-i jami‘

was largely incidental for Juzjani, its entrance merely serving as the stage
for display of part of the plunder from Ajmer, which probably reached
Firuzkuh c. ah 590–591/1194–1195 ce (cf. Volume II). At the same time,
even though Juzjani focused more on the Shansabanis’ qasr or royal palace
upon Baz Kushk-i Sultan than any other structure, the palace complex’s
precise dates of construction remain unclear, as it was already in place

136 Iran to India
when the Ajmer plunder arrived. Despite – or perhaps because of – the

considerable time that both Juzjani and his father Maulana Siraj-i Minhaj
spent at Firuzkuh, the author provided no precise description of the site.
In fact, his mentions of the city seem to “verge on the incredible” and
“result in a conflated account of a capital …” when compared with the
recent archaeological work at the site, as discussed further on.11
In light of the role played by architecture – particularly palatial
architecture – throughout the Persianate ecumene until the late eight-
eenth century,12 the dissonance between Tabaqat’s formulaic descriptions
of the site and its archaeology are, in fact, not surprising. Chapter 1,
argued that, for pre-modern power-seeking elites, architectural patronage
was an important means of labor mobilization, and the resulting structures
also served as monumental markers of territory and as effective tools for
the establishment of bases of socio-cultural power. In short, architectural
activity was instrumental in the creation of an empire, as further discussed
in Chapter 4.
But for pre-modern Persian authors, their patrons, and their audiences,
the palace complex was both a symbol of royalty as well as “a special place
as a nexus of power which join[ed] the human and the cosmic … a symbol
of something beyond the merely material.” The palace had a “rhetorical
function … through its affect, [which was] as essential as its residential,
administrative, productive, and ceremonial functions.”13 Thus, like the
poetic-literary, religious and other intellectual luminaries gracing royal
courts, patronage of palaces also provided one of the many manifesta-
tions of royal prerogative and power. Indeed, Juzjani’s descriptions of
Firuzkuh’s buildings contain many elements directly echoed in qasidas
(panegyric Persian poems) of the late eleventh–twelfth centuries, whose
metaphorical flourishes help us not so much to “see” palaces as perceive
what they meant.14
As might be expected, then, Firuzkuh’s archaeology does not corrobo-
rate Juzjani’s descriptions. Instead of a grand qasr commanding the center
of Firuzkuh, it is the preponderance of defensive, religious, and possibly
commercial buildings that has been archaeologically attested. It was par-
tially this textual–archaeological discrepancy, along with other historical
preconceptions, that had resulted in grave doubts about Jam-Firuzkuh as
the Shansabanis’ summer “capital.”15 But the work of Dr. David Thomas
and the Minaret of Jam Archaeological Project (MJAP) between 2003
and 2005 have further eroded lingering skepticism of Jam-Firuzkuh as the
Shansabanis’ sard-sir destination. Even with limited time and curtailed
resources, Thomas and his colleagues conducted excavations, surveys,
soundings and analyses – the last also encompassing the finds from the
staggering number of robber holes marking the site’s mountainous terrain.
Indeed, MJAP’s (ongoing) studies, together with continuing anthropo-
logical research on nomadic and transhumant populations, and the con-

sequent rethinking of archaeological methods for recovering the latter’s

physical traces, have helped to bring into focus a plausible picture, viz. a
mercantile settlement that was adopted by an increasingly powerful and
persistently transhumant elite as one of their seasonal encampments.
As is the case with most of the cultural geography of Afghanistan (see
also below), all discussions regarding Jam-Firuzkuh and its surroundings
in the mid- to late twelfth century must bear in mind at least three provi-
sos, two historical and one modern. The historical provisos emerge from
textual and archaeological indications: Juzjani mentioned that a flood
destroyed the Firuzkuh jami‘ during the reign of Sultan Ghiyath al-Din
Muhammad, but after his brother Mu‘izz al-Din’s successful India cam-
paigns in the 1190s16 – thus only a little more than two decades before the
Mongol campaigns into Afghanistan (1220s). This “major depositional
event” was documentable in the most recent archaeological exploration
of the area east of Firuzkuh’s famous minar – as discussed below, a zone
that yielded a probable mosque or prayer area (Figure 3.1).17 This natural
disaster apparently wrought extensive destruction on what was likely the
politico-religious focal point of the settlement, where other structures of
note could have existed – and noted by Juzjani – but were ultimately swept
away by the flood.18 And of course we cannot discount the Mongols: in
ah 617/1222 ce Firuzkuh underwent extensive destruction and large-scale
slaughter of its remaining inhabitants at the hands of the Mongol Ögedey
and his armies, so that much (if not all) of the city’s pre-Mongol architec-
tural fabric is no longer visible.19
Coupled with this compromised archaeological context is the forced
incompleteness of archaeological survey and excavation. Given the end-
lessly fluctuating security situation in central Afghanistan, and govern-
mental delays in issuing permits, the admirable work of Dr. Thomas and
the other MJAP members had to be left only partially finished as of 2005.
While he and his specialist colleagues in the field continue their analy-
ses and publication of the data already collected (including GIS studies,
ceramics, epigraphy, etc.; see below), many aspects of the site’s archae-
ological context remain unexplored. All the observations offered below,
then, must remain hypothetical, subject to modification depending on the
emergence of more archaeological and other data over time, which may
add detail to what is proposed here or open alternative understandings of
the site.
Nevertheless, MJAP’s explorations of Jam-Firuzkuh have supplied crit-
ical information regarding the erstwhile summer “capital,” which Tabaqat
lacked. The team’s survey of domestic architecture, concentrated on the
sloping west bank of the Jam Rud facing the famous minar, documented
structural remains of well-built walls of stone, together with combination
walls of stone and mud brick, some with plastered surfaces. The data
indicated a populous terraced settlement, though likely of relatively short

138 Iran to India
Figure 3.1 Jam-Firuzkuh: remains of paved floor of probable mosque located east of minar.
© Courtesy of David Thomas.

occupation. But the “concern for, and investment of resources in, the
aesthetic qualities of the buildings” signaled that a settled, rather than
transhumant or nomadic, population resided in these permanent dwell-
ings during much of the year. Furthermore, remnants of a large, arcaded
structure on the Hari Rud’s north bank were not for defensive purposes,
but rather “more reminiscent of the shop fronts in the traditional souqs of
the Islamic world.”20 The archaeological evidence thus far, then, supports
the idea that Firuzkuh was an already existing mercantile settlement –
Juzjani’s cosmopolitan Jewish merchant serving as its personification (cf.
Chapter 2) – with which the early Shansabanis established a mutually
profitable association, and which probably brought further growth in the
area’s settled and nomadic populations.
The one complex for which no specific archaeological evidence could
be found was Juzjani’s lofty and crenellated Kushk-i Sultan, supposedly
located atop a natural elevation (Pers. baz) in the center of Firuzkuh and
serving as the Shansabanis’ qasr. Instead, indications of elite occupation
were concentrated largely in defensive structures on surrounding prom-
ontories. The ruins known collectively as Qasr-i Zarafshan (Figures 3.2
and 3.3), for example, stretch along the north bank of the Hari Rud, the
most substantial remnant being a conical tower of mud bricks on a stone
footing. The tower is comparable with the defensive remains documented
in the south of Ghur (Chapter 2) and west of Bamiyan (see below), though
it lacks the plaster surfaces with molded or incised designs. However, the
Qasr’s local name of Arg-i Dukhtar-i Padshah – “Citadel of the Emperor’s
Daughter” – seems to carry a royal association.21 Farther above the Qasr
stands the “refuge” called Kuh-i Khara (“Black Mountain”?), which pro-
vided more certain signs of elite occupation: a large cistern lined on three
sides with baked bricks (Figure 3.4); turquoise-glazed brick fragments;
and a significant number of high-status ceramic sherds such as mina’i
ware. Although, the disturbance caused by numerous robber holes makes
it difficult to put forth definitive conclusions regarding the full range of
occupation at these primarily defensive structures.22
The Jam-Firuzkuh remains are not altogether unusual in light of the
Shansabanis’ continued transhumance and/or nomadism throughout the
twelfth century, consisting of seasonal movements between their summer
encampment or “capital” of Firuzkuh and its winter counterpart in
Zamindawar (cf. Chapter 2). Again, their contemporaries the Saljuqs offer
a parallel modus vivendi. The Saljuqs of Iran (c. 1038–1194) largely preferred
a textile palatial architecture in the form of luxurious royal tents, and at
times cloth enclosures (saraparda) incorporating architecture with a limited
footprint such as a kushk, a single- or two-story pavilion for viewing (or
being viewed).23 The name of Kushk-i Sultan for the Firuzkuh qasr could
very well have referred to just such a structure – swept away by the flood
of 1199 ce and/or the Mongols. Thus, Juzjani metaphorically – rather than

140 Iran to India
Figure 3.2 Jam-Firuzkuh: tower remains of Qasr-i Zarafshan. © Courtesy of

David Thomas.
literally – attributed multiple towers, belvederes, and arcaded promenades

to the complex, in line with the royal symbolism expected in textual
(and largely poetic) descriptions.24 In practical terms, the nomadic and/
or transhumant lifestyles of the Saljuqs and the Shansabanis would have

Figure 3.3 Jam-Firuzkuh: remains of defensive towers. © Courtesy of

David Thomas.
required ample pasturage, which was not easily found within traditional
urban or settled areas. (As discussed in Chapter 2, these requirements
were difficult to meet also at Firuzkuh.25) Their extensive encampments
likely included large textile enclosures (saraparda) for the elites, as well as

142 Iran to India
Figure 3.4 Jam-Firuzkuh: baked brick, plaster-lined cistern at Kuh-i Khara,

looking north. From Thomas 2018: fig. 5.24.
individual tents, and would have been pitched outside established towns
and cities.26 Furthermore, given the Saljuqs’ reliance on armies of allied
nomadic groups – increasingly true also for the Shansabanis as of the 1150s,
if not earlier – vast, open spaces beyond an urbs were a necessity as well.
In the end, even though the Shansabanis quite plausibly relied mainly
on textile palatial architecture, Firuzkuh’s defensive structures, where
elite occupation is also in evidence, were still essential: it is possible that
the Shansabanis provided the security necessary for the mercantile trade
mainly carried out by the region’s networks of Jewish communities,
who in turn helped to replenish the Shansabanis’ (eventually impe-
rial) coffers, forming a politico-economic loop not only increasing the
Shansabanis’ power but also integrating them further into Firuzkuh’s
very existence.27
MJAP’s exploration of the area immediately to the east of the famous
minar (discussed further in Chapter 4) unearthed an evidently extensive
flooring of baked bricks, with two shaft fragments of toppled columns
of the same material lying on the southern edge (Figure 3.1). The floor-
ing consisted of varying patterns, ranging from herringbone to large rec-
tangular bricks laid as paving – the column fragments were found atop
the latter – strongly hinting at the possibility of a large, columned mosque
associated with the minar. But MJAP’s exploration of the area in five
soundings along its northern perimeter (edged by the Hari Rud’s southern
bank), rather than overall excavation, precluded the collection of many
other valuable data – a lacuna hopefully to be remedied by future work.
In the meanwhile, the paved floor’s likely extent and dimensions cannot
be determined, nor those of any associated structure. Further, while frag-
ments of a baked-brick wall were found abutting the riverbank delineating
the northern wall of the possible mosque, it remains unclear whether a

qibla wall existed: if so, it would have abutted the minar on the latter’s
east.28 Notwithstanding the missing information, MJAP’s analyses of the
baked-brick flooring and other findings concluded that the encompassing
structure was in all likelihood a mosque, and one to which the Shansabanis
were “unlikely to have devoted fewer resources” than to the magnificent
minar adjacent to it.29 However, it is unknown whether the unearthed
structure was the first one in that location, perhaps having superseded
an earlier (smaller?) mosque and even an accompanying (less imposing?)
minar.
It is generally assumed that Jam-Firuzkuh’s minar (Figure 3.5) and
the newly found mosque abutting it were constructed in tandem in the
early 1170s, prior to which the “capital’s” Muslim population – including
its persistently transhumant Shansabani elite – would have prayed and
gathered in humbler, less monumental or even temporary structures.30
However, the probable absence of a Shansabani palatial complex per se (as
described above) would not preclude their patronage of other structures,
which, for a relatively new elite of humble origins, would have fulfilled
symbolically more potent functions than the archetypal palaces of estab-
lished urban rulers. Juzjani’s incidental mention of Firuzkuh’s jami and
its monumental portal (pishtaq), where some of the plundered spoils were
displayed upon their arrival from Ajmer (cf. above), implied only that the
grand structure was already in place by the mid-1190s; this did not neces-
sarily preclude the existence of an earlier structure at the site, perhaps one
that was subsequently expanded or even replaced.
The Shansabanis’ continued transhumance would seem to demand that
some more lasting sign of their presence at Firuzkuh – such as a markedly
central place for Islamic worship – serving as a steady reminder of their
commanding status, perhaps especially during the initial phases of their
dominance over the settled and relatively prosperous mercantile popula-
tion of Firuzkuh. Based on the Shansabanis’ own activity during the earlier
part of the twelfth century – discussed in Chapter 2 – it is worth consid-
ering that the place of prayer unearthed by MJAP was at least initiated in
the late 1140s–1150s, shortly after Qutb al-Din established himself as malik
al-jibal at Firuzkuh, and more than two decades before the construction of
the famous minar, which had a very pointed epigraphic program and com-
memorative function (cf. Chapters 4 and 5). The Shansabanis had already
patronized laborers with a localized, vernacular familiarity with Persianate
architectural culture when they commissioned the construction of the Da
Qadi minar. This structure might have been ensconced within a mosque
complex, and the site has tentatively been identified as Qala‘-yi Zarmurgh,
their sard-sir “capital” prior to Firuzkuh (as argued in Chapter 2). But by
the mid-twelfth century, the Shansabanis’ prominence and resources had
risen considerably, allowing them access to skilled laborers who were au
courant of the broader, cosmopolitan Persianate architectural trends. These

144 Iran to India
Figure 3.5 Jam: minaret of Jam (c. 1180) [sic], general view. © Photograph
building conventions had been shaped throughout the previous two cen-
turies by the robust patronage of the Saljuqs and the military aristocracy of
the Ghaznavids. Thus, increased means and access to the necessary skilled
workers meant that the Shansabanis could commission the whole range
of architectural forms within the former’s repertory, including mosques,
minars, madrasas, ‘idgahs, and a variety of other buildings.
Indeed, the well-known Jam-Firuzkuh minar (Figure 3.5), discussed
in further detail in Chapter 4, may provide an iconographic hint of the
prior existence of an ‘idgah if not a mosque – neighboring the minar. In
her re-examination of the minar, archaeologist and epigraphist Janine
Sourdel-Thomine painstakingly analyzed its epigraphic and decorative
program. While the significance and uniqueness of the entirety of Surat
Maryam (Qur’an XIX) on the minar’s lower shaft is addressed in the
following chapter, here it is useful to highlight the low-relief motif of a
blind arch on its east–west axis (Figure 3.6): nestled among the complex
web of Qur’anic and historical inscriptions, and geometrical and vegetal

Figure 3.6 Detail of Figure 3.4: Jam, minaret, view of mihrab relief on east face.
ornamentation, a towering pointed-arch blind niche distinguishes the

eastern face of the minar from the other surfaces. In fact, the most impor-
tant verses of Surat Maryam (Q. XIX: 35–41) were ingeniously placed at
the apex of this niche motif, which, together with its qibla orientation,
strongly points to its function as a mihrab.31
Sourdel-Thomine cautioned against considering the minar’s orna-
mented niche as “a secondary mihrab, but rather [as] adding an Islamic
element to the symbolic, religious and sacred quality of a monument with
marked cardinal orientations” (i.e., via the decorative programs of each
of the eight faces).32 Nevertheless, a mihrab on the minar would have
been redundant vis-à-vis an adjacent mosque’s qibla wall, and too close
to the latter to allow for separate worship. Given the labor and resources
invested in the minar – including the difficult spatial calculations required
to integrate the mihrab motif and its meaningful verses seamlessly within
the complex decorative program – it is safe to assume that one of the
principal aims of the overall œuvre was to be visible, if not also legible (cf.
Chapter 4). Thus, at the time of the minar’s construction, it is possible that
an unroofed prayer area or ‘idgah might have already been in place, even
marking the qibla by means of a low wall permitting visibility beyond.33
The eventual construction of the minar in ah 570/1174–1175 ce obviously
served multiple purposes, as discussed in the next chapter; a significant

146 Iran to India
one, however, could have been the monumental marking of the direction
of prayer.
One final point awaits explication. Juzjani’s mention of the Firuzkuh
jami‘s pishtaq (cf. above) hints at a monumental structure rather than an
open space for prayer, probably in the same location to the east of the
minar. This grand, most likely enclosed, building was evidently in situ
by the mid-1190s, in time to display some of the Ajmer plunder. Such a
mosque and its surely impressive qibla wall would have eventually blocked
the view of the mihrab on the minar’s eastern face, seeming to replace it
with the mosque’s qibla wall. This blatant subordination of the minar’s
mihrab could have resulted from the gradually fraying alliance between
the Shansabani sultans and the Karramiyya, whose doctrinal sway over
the region had led to the unusual choice of the entirety of Surat Maryam
(Q. XIX) on the minar’s surface (cf. Chapter 4). But, by the 1170s and
afterward, the Shansabanis’ increasingly cosmopolitan ambitions required
distance from the Karramiyya’s admittedly reduced, localized appeal. It
is worth considering, then, that a new masjid-i jami‘ was constructed
sometime after the minar’s completion in 1174–1175 ce and before Juzjani’s
mention of the jami‘ pishtaq in the context of events occurring in 1194–1195
ce. This architectural gesture would have subtly but firmly sidelined the
Karramiyya, while still maintaining the minar’s symbolism as the com-
memoration of the definitive Shansabani conquest of Ghazna (see esp.
Chapter 4).
The Shansabanis’ continued transhumance and their use of Jam-Firuzkuh
as a sard-sir encampment helps to explain the absence of a permanent qasr,
the role of palace amply fulfilled by “textile architecture” possibly in com-
bination with small immoveable structures such as kushks. Nevertheless, it
might be expected that these ambitious elites made an early and enduring
statement of their newfound dominance over Firuzkuh, an already exist-
ing settlement with established commercial networks. Prior to the famous
minar, expending resources on the construction of a large place of prayer
or ‘idgah would unequivocally declare their religio-political identity, con-
comitant with their commanding status, in the language of imperialism
then current in the region: the strongly Persianate Islam of the Saljuqs and
Ghaznavids. Prior to ah 570/1174–1175 ce and the erection of the famous
minar, it is difficult to imagine how else the transhumant Shansabani elite
would have made their presence felt at their sard-sir “capital” during their
migratory absence for much of the year, if it were not for an enduring
monument like a place of Islamic worship.
Ultimately, Jam-Firuzkuh makes a compelling case for an entirely dif-
ferent way of understanding the import of Persian textual sources in the
study of something as seemingly indelible as architecture. Pre-modern
Persian authors relied on a repertory of visual and poetic symbolism pri-
marily derived from a sedentist royal court model, which did not concord

Figure 3.7 Bamiyan: north cliff, eastern end, with 35-m Buddha (fourth–fifth
century), distant view. © Photograph Josephine Powell c. 1960. Harvard Fine Arts
Library, Special Collections.
with the inner workings of a new elite continuing their seasonal transhu-
mance. This urban court model, then, should be considered an even greater
abstraction – rather than falsification – of the locus of power particularly in
the case of elites from obscure, nomadic or transhumant origins.34 With
the increasingly refined archaeological work underway on empires emerg-
ing from nomadic confederacies, the juxtaposition of textual descriptions
and archaeological finds is both increasingly imperative and, thankfully,
possible.
Bamiyan
The UNESCO world heritage site of Bamiyan (Figure 3.7) looms large
in the contemporary imagination. It is hard to think of Afghanistan or
Buddhism without evoking Bamiyan – the mind’s eye of the modern

148 Iran to India
world arrested as much by its Buddhist monuments, as by the destruction

of this magnificence in the name of a supposedly retrogressive Islamic
iconoclasm.35 In light of Bamiyan’s modern notoriety and its cultural and
political resonances, its true physical presence is commonly collapsed into
the narrow area of UNESCO designation. Bamiyan’s overall importance
seems towering enough to be timeless, so that the basin’s actual history and
geographical extent fade into the hazy mists of memory and imagination.
The historical town of Bamiyan had in fact sprung up due to its command
of a basin extending roughly 50 km east–west, with a maximum width of
15 km. It occupied the perfect position for travelers as they embarked on
or concluded journeys through difficult mountain passes, which were and
continue to be dangerous yet unavoidable challenges in traveling between
Central Asia, Bactria, and northwestern India. The principal Bamiyan
valley constituted “the great capital of a royal kingdom” according to the
seventh-century Chinese pilgrim Hsuan Tsang (602–644 ce).36 But a more
accurate historical conception of the locale would encompass the basin as
a whole, including the associated valleys of Funduqistan, Fuladi, Kakrak,
and Nigar (Figure 3.8), and the traversable passages to its west extending
ultimately into Ghur.37 Furthermore, if the remarkable large-scale artistic
and architectural patronage at the site of Bamiyan is any indication, its
transregional significance came into being no earlier than the mid- to later
sixth through tenth centuries ce (see below).
It is true that the location of Bamiyan northwest of Kabul, essentially
on the India axis from Balkh and Qunduz (via Kabul and Gardiz), assured
the area some traffic from the early centuries ce as the combined result
of commerce and Buddhist proselytism. To wit, archaeological explora-
tions of Bamiyan’s valley floor have recovered the remains of freestanding
structures, most likely Buddhist monasteries, whose occupation has been
assigned to the third through fifth centuries ce. Then a century-long lapse
intervened, though afterward the monasteries were reoccupied, and con-
tinuously so from the sixth through the late tenth centuries ce.
However, the dates of Bamiyan’s two famous colossal Buddhas, once
standing within painted niches in the principal valley’s cliff face, do not
span the entire period of the valley’s domination by Buddhist monas-
ticism; rather, they were made during a specific historical moment. Art
historians have dated these monumental works to c.550–600 ce, based on
stylistic comparisons of the Buddhas’ sculpted forms with more securely
dated works. Additionally, recent 14C analyses of surviving organic mate-
rials from the Buddha sculptures verified these conclusions.38 The remark-
ably abundant, consistently high-quality and large-scale artistic patronage
throughout Bamiyan’s valleys as of the later sixth century ce arguably
signified a greater concentration of economic and political resources in the
area as well.
The Bamiyan basin’s increasing prominence in the mid-first millennium

ce was not tied to a growth

in adherence to Buddhism
per se, but rather to a pivotal
recrafting of political alli-
ances in the region. After the
defeat of the Hephthalites
at the battle of Gol-Zarriun
(mid-sixth century) pri-
marily at the hands of the
“Chinese” or Western
Turks, the ever-present
Sasanian shahanshahs even-
tually allied with the latter
in order to maintain some
control of their eastern ter-
ritories. Aside from this new
regional power structure,
however, it remains unclear
whether the previously dom-
inant Hephthalites were
completely displaced, or still
present in regional pockets.
The interrelated dynasties of
Kabul and Zabul – the latter
being the Rutbils of the early
Arabic sources (see Chapter
2, and below) – and the
Shar of Bamiyan all pledged
loyalty to the Yabghu presid-
ing over the Western Turks’ Figure 3.8 The Bamiyan basin. From Baker and Allchin
territories south of the Oxus. 1996: map 3.
But it is debated whether the
local rulers were themselves Western Turk or Hephthalite, and whether
they relied for regional administration upon a newly created echelon of
Turks, or a lingering Hephthalite “aristocracy.”39
Notwithstanding the still unknown aspects of the region’s political
landscape during the mid-first millennium ce, it is evident that the rise
in Bamiyan’s fortunes coincided with the rise of the Western Turks in
Tukharistan (Bamiyan and northward). We saw that, during the early cen-
turies ce, Bamiyan had served as only a middling way station when small
Buddhist monasteries facilitated and benefited from the trickle of com-
mercial and other traffic. By c. 600 ce, however, Western Turk interven-
tions had transformed the Bamiyan basin into a cosmopolitan but largely
Buddhist commercial powerhouse, most vividly attested by its magnificent

150 Iran to India
artistic production, wherein “artists … continued to have access to the

highest quality materials [and] worked with iconographic themes related
to religious practices in contemporaneous Central Asia”.40
The new rulers and their loyal vassals and/or kinsmen were all prag-
matically ecumenical: in order to reconfigure commerce to their advan-
tage, they emphasized the north–south routes through Bamiyan, largely
displacing the east–west “trans-Gandhara” corridor as the primary access
connecting India and Central Asia. Already an established Buddhist site,
they might have further enforced passage through Bamiyan by adopting it
as the dynastic center of the newly ascendant Khaganate, while its political
center or regional capital was probably Qunduz.41 A dynastic imprimatur
would permit the Western Turks to profit greatly from Bamiyan’s traffic
of people (including merchants), goods, and ideas on the north–south
trajectory through their territories.42 Inadvertently or not, the Khaganate
also encouraged – perhaps patronized – the institutions and practices of
Buddhism already established in the area, even if these lay outside the
Western Turks’ ritual associations (cf. Chapter 2 and below).
The five contiguous valleys comprising the core area around Bamiyan
(Figure 3.8) appear to have been primarily dedicated to Buddhist monasti-
cism and dotted with artificial and natural grottoes, all domiciling Buddhist
rituals and monastic residences.43 However, political and administrative
control of the area, necessarily monitoring access along the various routes
in and out of the basin, was concentrated in hilltop complexes such as the
well documented Shahr-i Zuhak (Figures 3.9–3.11) and the spectacular,
recently re-surveyed Shahr-i Ghulghula (Figures I.22, 3.12, 3.13). Indeed,
Bamiyan and its immediate areas to the north and west are replete with
remains of rambling complexes large and small, usually atop high prom-
ontories, seeming to form a geographic, climatic, and even architectural
continuum with the Shansabanis’ home region of Ghur immediately to
the west and south (cf. Chapter 2).44 The mid-twentieth-century surveys
by Marc Le Berre (published in the 1970s–1980s); subsequent surveys,
documentation and analysis by Z. Tarzi (esp. 1977, 2007); and, finally,
excavations by P. Baker and R. Allchin (1991) at Shahr-i Zuhak have been
invaluable for obtaining a glimpse of the Bamiyan basin’s non-religious
architecture. This corpus, together with the Buddhist monasteries and
other religious sites,45 enables an overall apprehension of the region’s
pre-Islamic architectural culture, its continuation into later periods, and
its commonality with building practices in other regions of Afghanistan.
Shahr-i Zuhak (Figures 3.9–3.11) as the better-studied hilltop site in
the Bamiyan basin serves as an example of the area’s architectural devel-
opments. Archaeologically, the initial foundation of Shahr-i Zuhak and
nearby hilltop complexes has been attributed to the Sasanian–Hephthalite
period, up to c. 500 ce, though phases of occupation at most Bamiyan sites
continued for several centuries afterward. With the rise of the Western

Figure 3.9 Shahr-i Zohak (fifth–thirteenth century), general view, triangular

plateau with fortifications, from south. © Photograph Josephine Powell c. 1960.
Turks in the sixth century and the consequent shifting of political alli-
ances, a later phase of building apparently took place outside the core area
of Bamiyan in the secondary valleys. Nevertheless, virtually all complexes
datable to the pre-Islamic centuries shared broad characteristics.46 It should
also be emphasized that, while it is tempting to assign an exclusively military
function to the complexes atop Bamiyan’s rocky promontories – an inter-
pretation seemingly justified by the ongoing politico-military power-plays
in the region – close examination of the remains themselves indicates
that their uses varied, and that they acquired defensive and other military
functions only over time.47
Shahr-i Zuhak and other complexes consisted of brick walls atop foot-
ings of stone, or stone-rubble masonry (Figure 3.9) – a building method
also encountered in central through southern Ghur. Most complexes had
corner towers or bastions with circular or angular plans (Figures 3.9 and
3.10), their elevations exhibiting equally familiar geometric patterns. But

152 Iran to India
Figure 3.10 Shahr-i Zohak (fifth–thirteenth century), general view, crenellated

tower and walls. © Photograph Josephine Powell c. 1960. Harvard Fine Arts
in Ghur the decorations were pressed, incised into or molded from wet
plaster, while in Bamiyan they were made out of the brickwork itself (Figure
3.10). As has been noted for the architectural remains in central through
southern Ghur, in the Bamiyan region also, the degree and complexity of
exterior surface decoration underscored the possibility of non-defensive
functions, at least for some phases of the structures’ existences.48
The Bamiyan structures’ seemingly better state of preservation – or
perhaps, simply more available documentation – than their Ghuri coun-
terparts permits a clearer understanding of interior spaces as well: the
non-intersecting vaults and elliptical domes (Figure 3.11) are indicative
of pre-Islamic Central Asian building practices.49 Finally, in situ interior
mural paintings were documented particularly in the structures dotting the
routes of access to the west of Bamiyan (Figure 3.18).50 Altogether, these
data help to identify an architectural culture with minor regional varia-
tions encompassing large swathes of Afghanistan during the pre-Islamic

Figure 3.11 Shahr-i Zohak (fifth–thirteenth century), fortress ruins, interior.

corridor. © Photograph Josephine Powell c. 1960. Harvard Fine Arts Library,
periods of Sasanian, Hephthalite, and Western Turkic dominance – that

is, the sixth through tenth centuries ce – evidenced from Balkh through
Bamiyan and Ghur, and likely extending into Sistan as well.51
With the enduring shift in routes of travel and commerce to foster
Bamiyan’s position as a regional entrepôt, the area continued to be some-
thing of a valuable territorial possession for local powers even after the
Sasanians’ and Western Turks’ respective and gradual disintegrations. To
be sure, the relentless ‘Umayyad and ‘Abbasid incursions into Afghanistan
had disrupted political stability from the end of the seventh through
eighth centuries, also causing a noticeable diminution in the volume of
traffic through Bamiyan. Eventually, the Sistani Saffarids’ and ultimately
the Samanids’ successes (i.e., further erosion of Turkic power) restored
some calm, though the locale was no longer the great cosmopolitan center
of the preceding two centuries. Nevertheless, Bamiyan and contiguous
areas passed into the hands of the Samanids’ ghulam general Sabuktigin

154 Iran to India
(d. ah 389/999 ce), father of Sultan Mahmud of Ghazna (r. ah 389–421/

999–1030 ce), and eventually served as a Ghaznavid mint city.52 Bamiyan
remained in Ghaznavid possession until the mid-twelfth-century rise of
the Shansabanis.
According to Juzjani, shortly after Ala al-Din Husain Jahan-suz’s enraged
destruction of Ghazna and Bust-Lashkari Bazar in ah 545/1150 ce, he “turned
toward other conquests and captured the regions of Bamiyan and Tukharistan.”
Here, he installed his eldest brother Fakhr al-Din Mas‘ud, whose descendants
held the area until their rendition to the Khwarazm-shahs in the first decade
of the thirteenth century. But ‘Ala’ al-Din’s initial annexation was evidently
not definitive, as he had to return c. ah 548/1153–1154 ce to “[bring] under
his sway the districts of Bamiyan and Tukharistan,” along with the south-
ern areas around Bust (Zamindawar). This would have occurred upon his
release after several months in Saljuq captivity, during which time “a group
of amirs and dignitaries from the mountains of Ghur … and a group of
seditious persons from Kashi” not only put Nasir al-Din from Madin (‘Ala’
al-Din’s nephew) on the throne of Firuzkuh, they also “expropriated the royal
treasury and property.”53 It is possible that in suppressing this rebellion, ‘Ala’
al-Din had to reinstate control over the eastern tracts, attempting to declare
the paramountcy specifically of the Firuzkuh Shansabanis (cf. Chapter 1).
Despite these political vagaries, however, by all appearances Bamiyan and its
surrounding valleys continued to comprise a notable mercantile–agricultural
complex in the twelfth century. In this respect, it was surely not unlike
Jam-Firuzkuh; in fact, given Bamiyan’s much longer commercial history and
more advantageous location, the latter was at least as prosperous as its western
Shansabani counterpart, and probably more so.
As set forth in Chapter 1, cultural patronage – encompassing poetic, lit-
erary, and architectural creations – was a fundamental protocol of courtly
life throughout the Persianate world, emerging from the politico-economic
importance of cultural production. The Bamiyan Shansabanis came
to be notable cultural patrons, even rivaling their Firuzkuh cousins.54
For example, the well-known littérateur Nizami ‘Aruzi al-Samarqandi
(fl. c. 1100–1150s ce) was in Saljuq service during the first half of the twelfth
century, but later entered ‘Ala’ al-Din Husain’s court, which was presum-
ably at Firuzkuh. However, he ultimately dedicated his Chahar Maqala to
Husam al-Din ‘Ali Abu al-Hasan, a Shansabani prince of Bamiyan (Fakhr
al-Din Mas‘ud’s youngest son), indicating that he was probably lured to
the Bamiyan court by the greater benefits offered to him.55 Fakhr al-Din
Mas‘ud’s grandson, Sultan Baha’ al-Din of Bamiyan (r. ah 588–602/1192–
1206 ce), was apparently an even more eminent patron of letters and
architecture: this sultan patronized Fakhr al-Din ibn ‘Umar al-Razi (c.
ah 544–606/1149–1210 ce), one of the greatest philosophers and theolo-
gians after al-Ghazali (d. ah 504/1111 ce), occasioning al-Razi’s dedication
of Risala-yi Baha’iyya (no longer extant) to him.56 Juzjani additionally

recounted (at some length) that his own father Maulana Siraj-i Minhaj was
subject to Baha’ al-Din’s repeated entreaties to leave Firuzkuh and join the
Bamiyan court.57 Maulana finally acquiesced – in fact, “without the per-
mission of Sultan Ghiyath al-Din” – for which Baha’ al-Din rewarded him
with two madrasas “[having] grants (of land) and plentiful benefactions.”58
All the more surprising, then, is the apparent paucity in traces of
newly constructed twelfth-century architecture in the Bamiyan region.
Juzjani made note of amply endowed madrasas during the reign of Sultan
Baha’ al-Din – though without the embellishments of his description
of Firuzkuh’s Baz Kushk-i Sultan. Moreover, parallel to what we saw
at Firuzkuh, circumstantial evidence supports the assumption that the
Bamiyan rulers – at the very least Baha’ al-Din – would have patronized
civic and religious architecture, and also palatial complexes (cf. below).
But few remains can be dated exclusively to the late twelfth century. Of
course, we must bear in mind that subsequent intervention in a landscape
like Bamiyan – more heavily traversed than Ghur – would cause substan-
tial changes in and erasures of historical traces over almost a millennium.
The Mongol retribution visited upon Bamiyan (1221–1222 ce) alone could
have done much of the job, given the Chinggisids’ extraction of dispro-
portionate penalties from locales simply resisting their onslaught, not to
mention the extermination of entire populations in return for any deaths
of Chinggis Khan’s family members and associates in the course of the
sieges.59 Nevertheless, there are still surprisingly few indications of archi-
tecture newly built in the second half of the twelfth century.
By contrast, evidence of the reoccupation and repurposing of various
sites throughout the Bamiyan basin has survived the effects of time. Based
on the study of ceramic assemblages from several of the latter sites, their
reoccupation has been principally dated to the later twelfth-century
Shansabani command of the region.60 The excavations at Shahr-i Zuhak
(Figures 3.9 and 3.10), along with surface surveys of the nearby hilltop
complexes of Sarkhushak (Figure 3.23), and recent excavations at Shahr-i
Ghulghula (Figure 3.12) and in the core Bamiyan valleys, can be considered
together with the documentation of numerous historical remains on the
routes of passage to the west of Bamiyan (Figures 3.14–3.18). As it stands,
the available evidence seems to indicate a shifting Shansabani presence in
the region, quite plausibly mapping onto their consolidation of control
over the valley and its surroundings, and adapting to the defensive needs
of these newly acquired territories (see below). As we know, Bamiyan ulti-
mately served as the seat of a collateral and competing Shansabani lineage,
achieving variable success vis-à-vis its Firuzkuhi cousins.
Archaeological explorations of the Hindu Kush brought to light at
least ten advantageously perched complexes along Bamiyan’s western
access routes (Figures 3.14–3.18).61 Further documentation and – ideally –
excavation at least at some of these sites would be required for beginning

156 Iran to India
Figure 3.12 Shahr-i Gholghola, general view, citadel (eleventh–thirteenth

century). © Photograph Josephine Powell c. 1960. Harvard Fine Arts Library,
to attain chronological and functional understandings of the many ruins

atop these western promontories. Nevertheless, for now it is plausible that
initial construction in these areas came about sometime in the pre-Islamic
centuries, possibly during the proposed expansion of architectural activity
into the surrounding valleys as the Western Turks extended their reach
southward throughout Tukharistan and consolidated their presence in
Bamiyan (see above).
The surface decorations on many structural remains (Figure 3.15) have
been described as “characteristic Ghurid geometrical pattern[s] composed
of lines of incised triangles.” But given the enduring use of similar archi-
tectural decorations even into the nineteenth and twentieth centuries,
they are far from diagnostic of chronology. Indeed, the complexes’ struc-
tural techniques (Figures 3.16 and 3.17) as well as the remnants of interior
painted decorations (Figure 3.18) are instead indicative of initial construc-
tion in the pre-Islamic centuries of the mid-first millennium ce.62 As was
the case with most of the other hilltop complexes throughout the Bamiyan
region, these western sites were in all likelihood also reused and possibly
repurposed throughout the following centuries, making archaeological
excavation and analysis all the more necessary for precise dating.63
Another complex indicating twelfth-century occupation is Barfak II,
also known as Danestama, located to the northeast of Bamiyan on the right
bank of the Surkhab River (Figures 3.19–3.22). While the overall site com-

prised approximately seventy-five

grottoes and caves in the foothills
surrounding a group of structures,
a free-standing building among the
latter interests us here. This brick
structure with rubble footing was
square on plan, having near-circular
towers diagonally addorsed at the
corners, and hemispherical towers
centered in three of the four walls
(Figures 3.19 and 3.20). Probably
double-storied, it was configured
around a courtyard with four ivans,
the one on the southeast also serving
as a monumental entrance. Adjacent
to this entrance on its north was a
corner chamber with a mihrab in its
western wall (Figure 3.21). Though
missing the top, the remaining
niche had a flat (rather than hem-
ispherical) back wall, whose stucco
revetment was replete with inter-
laced hexagons and six-pointed stars
(Figure 3.22). Each of the shapes was
filled in with stylized floral, vegetal,
and circular motifs, in turn pierced
with numerous holes to create a
visual play of voids and solids, light
and shadow. Although M. Le Berre
expressed surprise at finding “in a
Figure 3.13 Shahr-i Gholghola, recovered wooden
monument of the Islamic period
door leaves (whereabouts unknown). From Baker and
the half-wall of diaper-masonry so Allchin 1996: 31.
characteristic of the Buddhist mon-
uments of Afghanistan and the pre-Islamic monuments of the central
Hindu Kush,” archaeologists have favored an eleventh- or twelfth-century
date for the structure as a whole, based on the presence of this mihrab and
its stucco decoration.64
Given its proximity to the major Buddhist and possibly Western Turk
dynastic center of Bamiyan itself (cf. above), and in light of the abun-
dance of surviving pre-Islamic structures throughout the area, it is not
impossible that Danestama was originally a pre-Islamic site, repurposed
during the twelfth century. We know, of course, that the continuation of
building practices despite political changes was consistent with the nature
of architectural cultures (see Introduction), so that brick walls upon rubble

158 Iran to India
Figure 3.14 Yakaulang, Saighan, and Kahmard Districts, western Bamiyan

Province, Afghanistan. From Lee 2006: map 1.
Figure 3.15 Qala‘-yi Gauhargin, western Yakaulang District, Bamiyan Province:

south-east side. From Lee 2006: fig. 3.
footings by themselves do not indicate pre-Islamic construction. But the

presence of pre-Islamic pottery fragments, even in lesser quantities than
Islamic sherds, signal possible occupation of the site prior to the proposed
twelfth-century date of the mihrab and its stucco revetment.65 Furthermore,
the overall plan of the structure, viz. four ivans around a central courtyard

Figure 3.16 Bamiyan vicinity, structural remains. From Le Berre 1987: pl. 87d.
and corner bastions, was commonplace among the free-standing Buddhist

monastic complexes of Central Asia, particularly of the fifth–eighth cen-
turies, and also eminently usable for Islamic-period institutions such as
madrasas, palatial structures, and even fortified military garrisons. Finally,
it is worth recalling that the region’s great commercial prosperity and royal
presence resulted in lavish architectural (and artistic) patronage through-
out the second half of the first millennium ce, much of which continued
to be serviceable to newcomers establishing a politico-military presence
in the region. Indeed, like their predecessors, the Shansabanis had taken

160 Iran to India
Figure 3.17 Killigan, eastern Yakaulang District, Bamiyan Province: detail of

ornamentation in room on south wall. From Lee 2006: fig. 32.
advantage of other usable structures in Bamiyan, and Danestama could

have been another such act of architectural pragmatism.66
But not all Shansabani intervention in the Bamiyan region involved
pre-existing structures. At variance with Zuhak and possibly Danestama,
and the numerous ruins to the west of Bamiyan, the fortified hilltop site
of Sarkhushak (Figures 3.23–3.26) apparently had no pre-Islamic archi-
tectural fabric in its extensive remains. Located to Bamiyan’s north and
commanding a promontory at the confluence of the Bamiyan and Shibar
rivers (Figures 3.8 and 3.23), the site’s principal construction probably
took place during the Shansabani ascendancy in Bamiyan, or the second
half of the twelfth century, likely having later phases of occupation as
well.67 Documentation of the site identified the remains of two residential/
palatial structures68 – one with multiple interconnected stories sophisticat-
edly arranged on a gradient (Figure 3.24); one mosque, and a tomb with a
mihrab69 (Figure 3.25); a possible hammam; a quadrangular and apparently
fortified structure with circular corner bastions70; and about forty artificial
grottoes in a spur of the rocky promontory.
As with the other Bamiyan-area complexes, more systematic exploration
is also needed here. But it is still apparent that Sarkhushak’s mosque
and tomb architecture utilized domical profiles, three-pointed arches,

Figure 3.18 Chehel Burj, eastern Yakaulang District, Bamiyan Province:

unidentified ornamentation. From Lee 2006: fig. 24.
and lofty interior spaces – all architectural devices associable with the
larger Persianate-Islamic world. By contrast, its residential/palatial struc-
tures consisted of elliptical corridors (Figure 3.26) directly evoking the
pre-Islamic fabric of Zuhak, and exhibiting “horizontal squinches” and
corbeled ceilings apparently akin to Indic methods of construction.71 Thus,
Sarkhushak not only distinguished itself as one of the rare, newly commis-
sioned constructions quite possibly of Shansabani patronage in the main
Bamiyan valleys, it intriguingly reconciled the continuity of the regional
architectural culture with discernibly Islamic ritual structures, while inte-
grating Indic building practices as well.
The recent DAFA-led Franco-Afghan excavations at Shahr-i Ghulghula
(Figures I.22, 3.12 and 3.13) are germane at this juncture. These investiga-
tions have unearthed fascinating indications of more than one phase of

162 Iran to India
Figure 3.19 Barfak II, a.k.a. “Danestama,” c. 80 km northeast of Bamiyan: plan.

From Le Berre 1970: 47.
reuse and reoccupation at the site, although no apparent new construction

can thus far be attributed specifically to Shansabani presence in the area
during the later twelfth century.72 Such a refinement of architectural chro-
nology emerges from comparison of the Ghulghula remains with those of
Sarkhushak.
Like Shahr-i Zuhak (see above), Ghulghula was also constructed during
pre-Islamic centuries, but Zuhak exhibited comparatively little evidence
of later reoccupation vis-à-vis Ghulghula: for example, extremely few
Islamic-period glazed wares emerged from Zuhak, while Ghulghula’s
ceramic corpus was dominated by a truly impressive quantity and variety
of Islamic-period painted glazed wares with vegetal, geometric, and cal-
ligraphic or pseudo-calligraphic decoration, as well as fine glazed wares

Figure 3.20 Barfak II/“Danestama,” east side of central courtyard.

From Le Berre 1970: pl. IIa.
with molded decorative designs in shallow relief appliqué.73 In addition

to Ghulghula’s extensive ceramic finds, earlier documentation of the site
had already revealed two notable mosques: one with a qibla ivan over-
looking a large courtyard; and the other, a larger one probably with four
ivans around a courtyard.74 Finally, the rediscovery of intricately carved,
double-leaf wooden doors (Figure 3.13) strongly suggested Shansabani
presence here: the arabesque motifs in relief, contained within incised
geometric shapes, create an overall effect not unlike a door leaf recovered
during Jam-Firuzkuh’s surface explorations.75 The pre-Islamic structural
cores at Ghulghula, then, quite possibly received additional architectural
details during Shansabani occupation in the second half of the twelfth
century.
Franco-Afghan surveys and excavations on the western slopes of the
Ghulghula promontory in 2014–2015 revealed extensive historical remains
of palatial architecture, albeit deteriorated due to modern occupation and
modification. Nevertheless, elevations and plans of the older architectural
remnants – some extending up to 10 meters above ground – were iden-
tified and analyzed, resulting in close parallels to surviving Ghaznavid

164 Iran to India
Figure 3.21 Barfak II/“Danestama,” mihrab from southeast. From Le Berre 1970:
pl. IIIc.
palaces constructed and occupied throughout the eleventh–early twelfth

centuries. The elevations of Ghulghula’s building remains were also punc-
tuated with blind arcades, not dissimilar to the large Southern Palace at
Lashkari Bazar (see Chapter 5): in both cases, the walls most often showed
a large, blind central arch flanked on each side by three smaller ones.
Moreover, the traceable contours of historical walls highlighted a prefer-
ence for cruciform configurations, that is, central courtyards with alcoves
in the centers of the four sides (the four-ivan plan), creating intersecting
perpendicular axes.76 Such configurations were also in evidence at Lashkari
Bazar and, most noticeably, at the royal palace of Ghazna, which was until
recently attributed to the Ghaznavid sultan Mas‘ud III (r. ah 492–508/
1099–1115 ce), but is now thought to have had several prior phases of
construction and occupation.77
In comparison with Sarkhushak’s distinctively “hybrid” palatial
architecture – combining Indic and Islamic-Persianate building forms and
construction methods – Ghulghula’s remains appear to be largely, if not
exclusively, reminiscent of the Persianate palaces of Ghazna and Lashkari

Figure 3.22 Barfak II/“Danestama,” detail of stucco revetment in mihrab.

From Le Berre 1970: pl. IV.
Bazar.78 It could be proposed, then, that the site overall experienced at least
two documentable phases of reoccupation: Whatever pre-Islamic fabric
remained at the site was certainly visited by later builders of palatial struc-
tures, probably during the Ghaznavid command of the Bamiyan region
from the end of the tenth through early twelfth centuries. With the estab-
lishment of a Shansabani branch at Bamiyan c. 1150 ce, the Ghaznavid
palatial remains could have been reoccupied, obviating the need for new
construction (as we will see in Ghazna itself: cf. Chapter 4).

166 Iran to India
Figure 3.23 Sarkhushak, north of Bamiyan. From Le Berre 1987: pl. 116a.
Figure 3.24 Sarkhushak, Building A, plans and elevations. From Baker and
Allchin 1996: fig. 5.13.
The Shansabani occupation of sites throughout the Bamiyan region

can be understood at least partially in relation to this branch of the
Shansabanis’ continued monitoring of access to the core valleys. Moreover,
their probable reuse of westward hilltop sites can also be explained by the
contiguity of Ghur to the west, and the traffic between the two regions.

But the geo-political mapping of power

had shifted as the Shansabanis gained a
foothold in the Bamiyan basin: rather
than the area being an extension of
power radiating southward throughout
Tukharistan with late Hephthalite–
Western Turkic control as in the sixth
through ninth centuries (see above), by
the later twelfth century the Bamiyan
Shansabanis had to contend with disaf-
fected Saljuq governors, the encroach-
ing Oghüz, and the Qarakhitay, all
poised to approach from the north.
The Shansabani construction of
Sarkhushak – boasting both palatial as
well as possibly military structures, built
by the architecturally skilled labor avail-
able in the region – surely aided in the
vigilance against an entirely new gamut
of enemies.79 The possible reoccupation
of Danestama also falls in line with these
northern threats, lending further weight
to a military or at least multi-faceted
purpose for this complex.80
If indeed the palatial complexes at
Sarkhushak are attributed to Shansabani
patronage, they may present something
of a surprise given the archaeologically
attested absence of a permanent royal
residence at Firuzkuh (cf. above). But
the temptation to compare the two Figure 3.25 Sarkhushak, Building E, plan and
agnatic lineages, and to reduce them elevation. From Baker and Allchin 1996: fig. 5.18.
to a seemingly necessary but ultimately
artificial parity, can be avoided if we recall the great variety both within
and among transhumant/nomadic groups. The acknowledged symbiosis
among these multiple lifeways, effectively constituting the nomad–urban
continuum, was one in which the Shansabanis had participated since
pre-imperial times; in fact, it was arguably a factor in their very ascend-
ancy as a regional elite still preserving some form of transhumance (see
esp. Chapter 2).
Shansabani clansmen had already been exposed to a more sedentist life-
style, and some may have taken it up at an earlier stage than is historically
documentable. While Firuzkuh and the Hari Rud’s banks continued to
function as sard-sir for the Firuzkuh Shansabanis, the Bamiyan basin’s cul-

168 Iran to India
Figure 3.26 Sarkhushak, Building A, vault. From Baker and Allchin 1996: fig. 5.17.
tural milieu – defined in part by its centuries of cosmopolitan Buddhism –

quite likely demanded a more sedentist court structure of the Bamiyan
branch of Shansabanis. In light of their earlier exposure to settled lifestyles
and economies, the Bamiyan lineage’s transition to palace dwelling – also
in reoccupied contexts, such as Shahr-i Ghulghula – appears more cred-
ible. Thereafter, in the course of consolidating and maintaining power,
the Bamiyan Shansabanis appear to have reoccupied and commissioned
new palatial complexes, not unlike their Ghaznavid predecessors.81 In any
case, the end result was that, despite their agnatic connections, the two

Shansabani lineages functioned variably as elites and performed royalty

quite differently from each other, conforming to the demands of the
localized politico-economic and cultural circumstances of the areas respec-
tively constituting their disparate bases of power.
Tiginabad /Old Qandahar
Excavations to the south of the historical site generally referred to as

Old Qandahar, located 4 km west of modern Qandahar,82 were con-
ducted between 1974 and 1978 – the last to have been carried out there.
They unearthed a long and surprisingly fluctuating history at the site, a
mystery exacerbated by the varying names applied to the locale and its
vicinity from the Hellenistic period onward. Although “Tiginabad” and
“Old Qandahar” may appear anachronistic given these vagaries of naming,
they are used here to focus on the settlement and its surrounding area’s
relevance specifically for Shansabani expansion during the second half of
the twelfth century.83
The transit routes radiating southeastward from Herat, passing through
the regions of Zamindawar and al-Rukhkhaj and ultimately reaching India,
were as important as their northern counterparts toward Balkh, or again to
India via Bamiyan and Kabul. Thus, the plains of al-Rukhkhaj were punc-
tuated with settlements testifying to uninterrupted long-distance trade
from the pre-historic eras onward (see also Chapter 2). Old Qandahar
was among the locales benefiting from this continuous flow of people and
goods through the region. Moreover, the fates of the city and its nearby
Buddhist monastery were discernibly intertwined.
Looking down upon Old Qandahar and its plains was the famous
Qaitul Buddhist complex (Figure 3.27) comprising a vihara and tall stupa,
which still seems to command the vista from its prominence atop the
Qaitul ridge. The Buddhist complex was founded quite possibly as early as
the third century ce during the Mauryan presence in the region, and akin
to most other Buddhist sites closely associated with travel routes, it served
as a commercial node for the regularly passing traffic of merchants, pil-
grims, and others. The Qaitul monastic establishment was actively patron-
ized through the seventh or early eighth centuries ce, a span of time when
Old Qandahar down on the plains was also a major emporium. The pros-
perity of both the city and the Buddhist complex was surely augmented
by their proximity to Zamindawar, the regional garm-sir destination for
the Rutbils undertaking seasonal transhumance between Zabulistan and
al-Rukhkhaj-Zamindawar (see Chapter 2).
By the eighth century, however, the increasing conflagrations between
these rulers and the incoming ‘Umayyad and ‘Abbasid deputies disrupted
commerce and travel via the southern routes emerging from Herat. The

Figure 3.27 Qandahar, Afghanistan. Qaitul stupa and citadel of Old Qandahar. © Photo by
Warwick Ball, c. 1970s.
14/09/21 5:39 PM
disturbances took their toll on the Qaitul Buddhist monastic complex,

abandoned by the eighth century without any scope for revival.84 During
the same time frame, if not before, Old Qandahar had also contracted
to a shadow of its former self. By the mid-twelfth century and the likely
Shansabani presence there, the site’s status was even more uncertain, having
experienced a pronounced decline in settled mercantile and associated occu-
pation over several centuries, though nomadic populations likely continued
visitations to the area. Later eleventh- and twelfth-century Persian texts
mentioned a certain Tiginabad in the same region, probably referring to
Old Qandahar’s later iteration as a settlement of greatly reduced size.85 In
fact, Juzjani implicitly confirmed the identification of Old Qandahar as at
least in the vicinity of Tiginabad, if not the same settlement, in the context
of the Shansabanis’ sacking of Ghazna and Lashkari Bazar in ah 545/1150
ce.86 But it should be noted that Tiginabad/Old Qandahar’s diminution
in both size and importance between the eighth and fifteenth centuries is
remarkable given its location, which had rendered the entire area “at all
times in its history … a major frontier stronghold of immense strategic
value.”87
Given Tiginabad/Old Qandahar’s minor status by the mid-twelfth
century (and earlier), the reasons behind a Shansabani campaign and/or
presence there, along with the nature and duration of their stay during
initial expansion outside their ancestral Ghur, may seem less than appar-
ent.88 However, the skirmishes over the site between the Ghazna Yaminis
and the Shansabanis highlight not only its importance for the continuing
power struggle in the region, the Firuzkuh Shansabanis’ ultimate victory in
capturing it further reveal the politico-economic differences between this
branch and their Bamiyan cousins. Certainly, Tiginabad/Old Qandahar’s
location between Ghazna and Bust-Lashkari Bazar made it much more
strategically relevant to the Shansabanis of Firuzkuh than to their cousins
at Bamiyan. In ah 552/1158 ce,89 ‘Ala’ al-Din Husain once again confronted
Ghaznavid forces and wrested the locale, interrupting the Yamini transit
corridor between Ghazna and al-Rukkhaj-Zamindawar.90 Moreover, the
discernible post-conquest “footprint” at Tiginabad/Old Qandahar seems
to reflect a distinctly Firuzkuh Shansabani character, rather than one
evoking their palace-dwelling Bamiyan relatives.
The above-mentioned excavations at Tiginabad/Old Qandahar during
the 1970s unearthed an extensive cemetery – surprisingly vast, in fact, with
an extent surpassing the settlement itself.91 The cemetery contained burials
largely following Islamic prescriptions, that is, north–south placement of
unburned bodies with faces oriented toward Mecca – the qibla being in
a westerly direction from South Asia – within graves devoid of any goods
buried with the deceased. Among the burials, two large mounds, which
the archaeologists referred to as Mound A and Mound B, were visible and
even prominent features of the landscape calling for excavation.

172 Iran to India
Figure 3.28 Old Qandahar cemetery, Burial Mound A, octagonal platform looking
south. From Whitehouse 1976: pl. III.
Mound A (Figure 3.28) contained two sequential burials, each of a small

child, both oriented north–south with the faces turned west. The grave also
exhibited features seemingly outside strict Islamic burial norms, however: the
first body was placed within a small and vaulted, subterraneous mud-brick
crypt, marked above ground by a rubble and mortar octagonal enclosure
that was itself filled in to create a platform. Atop the first burial was that
of another child, still within the octagonal enclosure and platform. The
entirety of Mound A had been covered over with earth. The second nearby
burial, Mound B, revealed itself to be less elaborate upon excavation, but
still following similar protocols: the body of an adult and probably elderly
woman facing west was placed in a north–south trench grave, surrounded
by a rectangular enclosure, the whole contained within an earthen mound.92
Scholarly consensus has generally dated Mounds A and B to the
mid-twelfth century, more specifically associating both with the Shansabani
presence at Tiginabad/Old Qandahar. The strongest chronological indi-
cations for this dating have been the ceramic fragments, whose earliest
sequences fell within the twelfth century based on comparable finds from
Lashkari Bazar.93 Moreover, the combination of discernibly Islamic and
non-Islamic burial practices in the two graves – and others may yet come
to light – could be interpreted as making a stronger case for Shansabani
association rather than refuting it: We have noted in Chapter 2 that the
Shansabanis’ own cultural practices were rooted within the pre- or early
Islamic temporal strata of west–central Afghanistan, with continuations
or modifications over time. It is not impossible, then, that burial prac-
tices continuing those of their pre-imperial, and more nomadic and/or
transhumant forebears were still in place during these early phases of their
ascendancy outside historical Ghur.94

As for the impetuses behind the foray to Tiginabad/Old Qandahar

in the first place, and the Shansabani presence documentable there: it
is not unimaginable that, with Ghaznavid power gradually declining in
the region – particularly after ‘Ala’ al-Din’s much touted destruction
of Ghazna itself and Bust-Lashkari Bazar – the Firuzkuh Shansabanis
attempted to revive Tiginabad/Old Qandahar as a commercially and
strategically pivotal emporium once again. After all, the eastward transit
routes from Herat skirting Ghur to the south, passing through Tiginabad/
Old Qandahar upon entering and exiting India, continued to be just as
viable as their northern counterparts.
As with Bamiyan and the slate of new politico-military contenders
and rivals appearing on the horizon by the twelfth century (see above),
Tiginabad again proved itself a strategic point of military operations: the
Oghüz occupation of Ghazna signified yet another (nomadic) enemy in
the region certainly for the Firuzkuh Shansabanis, whom Shihab al-Din –
though subordinate to his older brother Shams al-Din, titled Ghiyath
al-Din as of ah 558/1163 ce – had nonetheless persistently tried to dis-
lodge throughout the 1160s. But it was only through a joint effort of both
brothers that the Oghüz were finally routed; as a result, the Firuzkuh-based
Shansabanis could claim definitive occupation of the erstwhile Yamini
capital in ah 569/1173–1174 ce (after prior brief stints there by ‘Ala’ al-Din
Husain – cf. Chapter 1). This victory accrued to the younger Shihab
al-Din, who was finally crowned sultan, receiving the surely long-awaited
title of Sultan Mu‘izz al-Din (cf. also Chapters 4 and 5). Tiginabad/Old
Qandahar likely served a strategic purpose for the Firuzkuh Shansabanis,
then, but barring rediscovery of additional evidence, it appears that
Shansabani presence there left behind only burials of a very few, albeit
seemingly important personages.95 It is, of course, quite possible that the
continuing lifestyle of transhumance of this branch of the Shansabani clan
might have left other, less conventional archaeological traces in the mate-
rial record, recovery of which would require surveying and even excavating
afresh at the historical city and in its vicinity.96
Multiple Shansabani Imperial Architectures
Compared with the available documentation of Ghur’s built remains

(cf. Chapter 2), the more extensive archaeological explorations of the
Bamiyan basin as well as Tiginabad/Old Qandahar permit several over-
arching observations. There are noteworthy material and historiographi-
cal parallels between the Shansabanis’ appanages in Ghur, Bamiyan, and
al-Rukkhaj-Zamindawar. The core areas of Bamiyan and Firuzkuh were in
fact central nodes in the discernibly differing cultural and economic geog-
raphies of eastern and western Afghanistan, while the Ghazna–Tiginabad/

174 Iran to India
Old Qandahar region in the south was ensconced within networks linking
the Indic and Iranian worlds since at least the early first millennium ce.97
Not only did these regions serve as stages upon which an obscure clan,
originating in the region’s nomad–urban continuum, coalesced as impe-
rial lineages; they also adumbrate the Shansabanis’ ingenious reuse of
inherited landscapes replete with the remains of rich and deep histories
(see also Chapter 1).
Indeed, all three areas must also be recuperated from the elisions and
silences of textual sources. Past authors’ virtual erasure of the Shansabanis’
origins in the transhumance/nomadism of Ghur – and everything else that
was erased alongside – have already been discussed (see Chapters 1 and 2).
But a similar silence surrounding the Shansabanis’ reuse and repurposing
of pre-Islamic sites and structures in the Bamiyan region is difficult to
ignore. Indeed, without the analysis of the surviving architectural remains
at Firuzkuh and Bamiyan, knowledge of Shansabani presences there would
be limited to the perfunctory textual mentions of the cultural patron-
age incumbent upon Perso-Islamicate rulers, with no intimation of their
remapping of the geo-political terrain as they contended with new adver-
saries. And without even the sparse archaeological data from Tiginabad/
Old Qandahar, insight into the impetuses behind the Shansabani cam-
paigns into al-Rukkhaj-Zamindawar would be altogether unrecoverable.
It bears reiteration that the Bamiyan Shansabanis in particular might
have conformed to their specific local requirements for maintaining power
by constructing palaces. This architectural patronage, arguably quite new
for a Shansabani clan, did not preclude reoccupation and repurposing of
a built environment, making the reuse in Bamiyan really a continuation
of the collective Shansabanis’ pre-imperial transhumant/nomadic prac-
tices (see Chapter 2). The simultaneity of these seemingly divergent ways
of shaping a landscape actually encapsulates the real complexity of the
nomad–urban continuum, wherein populations adapted to their chang-
ing circumstances with surprising alacrity. In their respective continuities
and adaptations, the Shansabani imprints at early Firuzkuh, Bamiyan,
and Tiginabad/Old Qandahar prefigure the subsequent, place-specific
empire-building that the lately established third branch – the Ghazna
Shansabanis – undertook throughout farther-flung conquests eastward to
the Indus and beyond (cf. esp. Chapter 6, and Volume II).
Furthermore, by the 1170s it could be said that there was no singu-
lar Shansabani elite, but rather separate and even competing agnatic line-
ages at Firuzkuh, Bamiyan, and eventually Ghazna (cf. Chapters 4 and 5).
The fragmentation of inheritance, customary among modern nomadic
groups (cf. Chapters 1 and 2), could in fact be suggestive of the increas-
ing politico-economic independence of these various Shansabani lineages.
Seniority, however, rested ostensibly with Ghiyath al-Din and the Firuzkuh
Shansabanis: this branch was part of a westward geo-historical momentum

reinforced by Firuzkuh’s mercantile economy and connectivity with the

cosmopolitan cultural geography of greater Khurasan via Herat, the age-old
commercial emporium and politico-cultural center, whence the routes of
communication branched out to the east and north (cf. Introduction and
Chapter 2).98 Firuzkuh’s westward links had distinguished its eponymous
appanage from that of Bamiyan since the days of ‘Ala’ al-Din Jahan-suz, for
example, in the many important military campaigns that could be under-
taken with a consistently replenished treasury, in addition to the marital
alliances with neighboring regions (cf. Chapter 4). A third imperial lineage
at Ghazna, initiated as of the mid-1170s with Shihab al-Din’s crowning as
Sultan Mu‘izz al-Din there, garnered great notoriety with their spectacular
victories in India (cf. Chapters 5 and 6, and Volume II).
All the principal and minor Shansabani lineages were interrelated and
also intermarried; but, as underscored here, each was increasingly inde-
pendent and adaptive to its respective base of power. Indeed, this very
responsiveness to particularized circumstances was a modus vivendi argu-
ably deriving from the Shansabanis’ earlier transhumance, or physical and
affective mobility between nomadic and sedentist lifeworlds. The respec-
tive Shansabani lineages established at Firuzkuh, Bamiyan, and eventually
Ghazna responded to and helped to shape their regional specificities, con-
stituting yet another glimpse of “the organization talent [sic], cosmopol-
itanism, pragmatism, adaptation, and integration ability of … nomadic
societies”.99
Notes
1. For a distillation of bibliographic references, see Ball 2019: 468.

2. Based on comparisons with firmly identified coinage and other factors (e.g.,
invocations of Baghdad khalifas), O’Neal (2016b) has attributed several pre-
viously unexamined coins of ‘Ala’ al-Din Husain to the Ghazna mint c. ah
555/1160 ce. This attribution would indicate that Ala al-Din again successfully
took the city and considered it a part of his domains before the Oghüz occu-
pation of the 1160s. Cf. infra in main text.
3. Remarkably, it appears that the reoccupation of Ghaznavid-patronized
palatial and other structures at Ghazna and Bust-Lashkari Bazar by Shihab
al-Din – properly called Mu‘izz al-Din after his installation and crowning at
Ghazna – and his ghulams was not inconsequential: the palatial architecture
attributable to the period of early Shansabani establishment in northern India
c.1192–1200 has much in common especially with precedents at Lashkari
Bazar, probably signaling the increasing habitation of built structures – rather
than the “textile architecture” of before – at least by the Ghazna branch of the
Shansabanis and their ghulams. See esp. Chapter 5.
4. See infra (main text and notes) for a discussion of Nizami ‘Aruzi’s Chahar
Maqala and its importance for contextual information regarding the Firuzkuh

176 Iran to India
and Bamiyan Shansabanis, and the significance of cultural production by

means of elite patronage.
5. See esp. Thomas’s (2018: 146–150) overview of the debates surrounding the
identification of Firuzkuh as the enigmatic Shansabani “capital”; cf. also
Chapter 2.
6. As discussed infra in the main text, doubts regarding precise locations
persist despite – again – O’Neal’s (2016a) excellent mapping of the known
Shansabani-associated toponyms. Cf. also O’Neal 2013: 16.
7. See Chapter 2, for Jam-Firuzkuh’s integration into regional commercial
networks, thanks in large part to the mercantile Jewish community estab-
lished there as of the eleventh century, prior to Shansabani presence. Cf.
also Juzjani, vol. I, 1963: 336 (trans. vol. I, 1881: 339); O’Neal, 2016a; Thomas
2018: 216.
8. See O’Neal 2016b for Wajiristan’s location; also Leshnik’s (1968: 47–48) dis-
cussion of Warshada in relation to Firuzkuh, etc.; and Thomas 2018: 39–41.
M. Ashtiany, the commentator on Baihaqi’s surviving text (most recently
translated by Bosworth), also confronted the puzzling mentions of place
names, noting that “the accounts of the Ghaznavid campaigns into Ghur
provide considerable onomastic and topographical information, whose sig-
nificance is unfortunately now obscure” (cf. Baihaqi [trans. Bosworth] vol.
III, 2011: 97–98). In his copious commentary on Juzjani’s text, Raverty (1881:
339, n. 8) noted other historical writers’ (though specifically citing only the
seventeenth-century Burhan-i Qati‘ ) mentions of Istiya, which was “the name
of one of the mountains … between Ghaznin and Hirat” – still too large a
swath of the country to identify the locality, probably a mountain fortress.
However, the apparent extension of the Istiya appanage to include Tiginabad
during Shihab al-Din’s (later Sultan Mu‘izz al-Din) possession of the latter
would indicate its location somewhere in southern Ghur, near the garm-sir of
Zamindawar and Tiginabad/Old Qandahar (cf. Bosworth [1977] 1992: 122;
O’Neal 2016a; and infra in main text). Furthermore, the location of Madin
could have been in either northeastern Ghur, within the ambit of Kashi, or in
the northwest near Gharjistan: the unsuccessful rebellion c. ah 547/ 1152–1153
ce and attempted occupation of Firuzkuh during ‘Ala’ al-Din’s “imprison-
ment” by the Saljuq Sultan Sanjar (r. ah 512–552/1118–1157 ce) – when he
actually served as the Sanjar’s boon companion – apparently originated near
Kashi, and the usurping claimant to the seat was “Malik Nasir al-Din Husain,
son of Muhammad of Madin, who was the son of the brother of ‘Ala’ al-Din.”
Cf. Juzjani vol. I, 1963: 348 (trans. vol. I, 1881: 361–362); see also below in
main text. It may be more likely, however, that Madin abutted Gharjistan, as
Saif al-Din Muhammad’s (r. ah 556–558/1161–1163 ce) campaign against the
Oghüz, who were encroaching from the northwest, took him to Gharjistan
and Madin. See Juzjani vol. I, 1963: 352 (trans. vol. I, 1881: 367 and n. 5).
9. Juzjani vol. I, 1963: 336, 337 (trans. vol. I, 1881: 339, 341). See Ball 2019: No.
468, for a brief site description and map, as well as an invaluable bibliography.
10. Juzjani vol. I, 1963: 375–376 (trans. vol. I, 1881: 403–404, 405). Even a cursory
comparison between Raverty’s nineteenth-century translation of this passage
and my own here makes apparent the need for a fresh examination and trans-

lation of Juzjani’s Tabaqat overall. The comparison may not reveal egregious
differences at first sight, but close scrutiny, particularly of the architectural
terms and their optimal translations, demonstrates the importance of Tabaqat
overall while also urging the text’s juxtaposition with other literary genres to
obtain its “hors-texte” significances.
11. Cf. Thomas 2018: 149, 151. It bears noting that scholars have similarly lamented
the lack of exact details in prose and poetry produced also at the Saljuq court
at Isfahan, which Durand-Guédy (2010: 79–80, 88–90) has posited as the
Saljuq summer capital, at least as of the reign of Malik Shah when the need
to “domesticate” the Turkmen nomads among Saljuq ranks came to the fore.
Nevertheless, where there is incidental and largely metaphorical–ceremonial
mention of architecture, it becomes apparent that “the Saljuq dynasty left its
mark on spaces of two kinds … the military and the religious,” rather than
the palatial (cf. ibid., 92–100).
12. It is noteworthy that the textual engagement with and utility of architecture,
particularly across the various genres of Arabic and Persian historical writing,
seemed to remain symbolic and metaphorical until the late early modern
period. Descriptions of the physical reality and even materiality of the built
environment apparently shifted only in late eighteenth–early nineteenth-
century historical writing: see esp. Khazeni 2018.
13. Cf. Meisami 2001: 21; and ibid., 42 (citation of Winter 1993: 39).
14. Meisami 2001: 22.
15. The debates surrounding the location of Firuzkuh and whether the minar
at Jam was indeed part of the Shansabani “summer capital” originated
during the later nineteenth-century Afghan Boundary Commission’s
information-gathering tours (esp. Holdich 1910: 223ff.). It is noteworthy that
many of the variant opinions were engendered by historically anachronistic
preconceptions regarding capitals (e.g., Leshnik 1968); processes and markers
of Islamization (e.g., Maricq and Wiet 1959); and the transhumance/nomadic
pastoralism of the Shansabanis (e.g., Wiet 1959; Herberg 1982). Vercellin
(1976) did, nonetheless, address some of these issues.
16. Cf. Juzjani vol. I, 1963: 375 (trans. vol. I, 1881: 404).
17. Thomas 2018: 160ff.
18. Pinder-Wilson (2001: 166) lamented that Juzjani did not specify
whether it was the Jam Rud or Hari Rud that flooded and overtook the
mosque, but seemed to spare the minar; MJAP’s invaluable archaeo-
logical work at the site all but definitively indicated that it was the Hari
Rud that broke its banks, since “fluvial deposits overlying the courtyard
paving in all the soundings [became] thinner to the west” (Thomas 2018:
173).
19. Cf. Thomas 2018: 152; also Juzjani vol. II, 1963: 127–128, 132–133 (trans. vol.
II, 1881: 1047–1048, 1057). There is archaeological evidence that Firuzkuh’s
settled population witnessed at least some destruction of households and
belongings, according to Thomas (2018: esp. 205). However, it should also
be noted that the continued transhumance of the Firuzkuh Shansabanis and,
quite possibly, the largely “textile architecture” of these elites would have
presented difficulties in recovery from the archaeological record even without

178 Iran to India
Mongol destruction. Cf. Durand-Guédy (2013a) for comparative examples

from Saljuq contexts throughout Iran; and also infra in main text.
20. Thomas 2018: 184, 190–191.
21. Noted by Herberg and Davary (1976: 61) and utilized by Thomas (2018:
174–176) in his analysis of the archaeological finds at Qasr-i Zarafshan.
22. Thomas 2018: 173–184.
23. For the Saljuq royal encampment at Isfahan, see esp. Durand-Guédy 2010:
93–101. Cf. also Durand-Guédy 2013a: 175–182 for a discussion of the kushk
incorporated into Saljuq encampments. The author also brought to light
intriguing discrepancies between various Persian textual passages referring to
the same royal residential complex, most notably the periodic substitution
of the term kushk with qasr (ibid., 153). Evidently, some historical Persian
authors deemed qasr, rather than kushk, to be more in keeping with the
unique role played by palaces as embodiments of royal power.
24. It cannot escape our notice that Juzjani largely relied on what was by the elev-
enth century not only the “basic repertoire of typical components of palace
descriptions,” but also recognizable rhetorical devices. The former included
manzar (belvedere), riwaq (gallery, columned promenade), and shuruf
(towers or crenellations); while the latter were exemplified by descriptions
of the unprecedentedness of the overall complex, symbolized in the inability
of the engineer/geometrician (muhandis) to have imagined or created such a
palace before. Cf. supra Juzjani’s description of Baz Kushk-i Sultan; and esp.
Meisami 2001: 24, 26, 28, 29.
25. According to D. Thomas (pers. comm., April 2018), it is difficult to identify
where, exactly, in the vicinity of Jam-Firuzkuh the space would have been
available to accommodate large encampments.
26. Recent scholarship on the Saljuqs and their building of empire rightly eschews
any essentialist idea of an innate preference for nomadic pastoralism ensuing
exclusively from their Turkic origins; rather, Durand-Guédy (2013a: 173) has
suggested that “the distribution of power [in the medieval Persianate world]
precluded them from living like urban Iranians.” Cf. Chapter 1; also Durand-
Guédy 2013b; Durand-Guédy 2018; Peacock 2015: 143ff., 172. For a discussion
of small architectural elements incorporated into Saljuq tent “palaces,” see
esp. Durand-Guédy 2013a: 154ff. Not only the early Saljuqs but also the later
Saljuq Sultan Malik Shah (r. ah 464–85/1072–92 e) – who “enjoyed unpar-
alleled authority during his long reign” – maintained a studied distance from
even their major urban conquests: Tüghril and Chaghrï remained in their
camp after their momentous conquest of Marv (c. 1035); Tüghril also decided
against settling in Baghdad after its occupation in ah 447/1055 ce; Malik
Shah preferred the environs of Isfahan; and, of course, Sanjar was frequently
encamped somewhere in the vast expanses surrounding Marv, Herat, Tus-
Nishapur – near which the twelfth-century Robat Sharaf may have provided
part of the royal encampment (Figures 1.10–1.13) – and Isfahan. Cf. Durand-
Guédy 2013b: 328–335; Peacock 2015: 143, 166–172.
27. Not all of the surveillance or defensive structures documented at Firuzkuh
proffered signs of elite occupation, nor were they all necessarily of the same
date. In 2003 MJAP re-examined ruined towers first documented by Herbert

and Davary (1976), to the east of the probable mosque (see infra in main
text) and minar on the north and south banks of the Hari Rud: with the
exception of one ruin, they were of weak construction and likely not for
defense but vigilance, seemingly designed for inward surveillance of the city
and its inhabitants, rather than focused outward toward the valleys granting
access to Jam-Firuzkuh from the east. Thus, Thomas (2018: 184) proposed
that these eastern towers might have been constructed during the tumultuous
pro-Karramiyya riots of 1199 (cf. Chapter 4), or after the Shansabanis’ fall and
Firuzkuh’s occupation by the Khwarazm-shahs, c. 1215.
28. For the wall remnants, see Le Berre, in Sourdel-Thomine 2004: 54–61, and
ibid., 95; also Thomas 2018: 159–160. The placement of the Jam-Firuzkuh
minar has been described as “idiosyncratic,” given its location on the qibla
side of the mosque (Flood 2005b: 538). Pinder-Wilson (2001: esp. n. 43)
noted that Saljuq minarets were frequently located at or near the north corner
of their mosques. Although, the Nayin jami‘s (tenth century) southeastern
minaret, and Barsiyan’s (eleventh–twelfth century) on the northwest rather
supports Hillenbrand’s (1994: 154) observation that “there seems to have been
no consistent practice governing the location of single minarets within the
mosque.” The location of Jam-Firuzkuh’s minar at the northwest corner of
the probable mosque/‘idgah-minar complex (as described infra in main text),
then, may actually indicate an overall adherence to Saljuq precedents, rather
than a deviation from them.
29. Dr. David Thomas, pers. comm., May 2018; cf. also Thomas 2018: 160–173.
30. Dr. David Thomas, pers. comm., May 2018; see also Flood 2005b: 538; Flood
2009: 95–96.
31. See Sourdel-Thomine 2004: 93–95, 149, 154; Flood 2005a: 276, 279–281; Flood
2005b: 538, 541; Flood 2009: 99; Lintz 2013: 91, 95–96. By consulting his own
dig books and calculating orientations via Google Earth, Dr. David Thomas
(pers. comm., July 2018) verified that the minar’s niche motif was oriented
to approximately 260°, thus plausibly functioning as a mihrab (from Jam,
Makka is 245°). The Karrami concordance with some Hanafi tenets (Zysow
2012: pts. I, ii, v) would have made the preferred Hanafi qibla in Central
Asia of due west (270°) acceptable also to the Karramiyya. For the custom-
ary qiblas of the Hanafis and Shafi‘is in Central Asia, see D. King (1982,
1995); and esp. Chapter 4. Moreover, Dr. Thomas 2005 documentation of a
modern “nomad mosque,” roughly outlined with stones and oriented to the
minar’s east face, indicated that the niche served as a mihrab in recent times
as well.
32. Sourdel-Thomine 2004: 95 (emphasis added).
33. Shokoohy and Shokoohy (1987: 129–132) proposed that the ‘idgah they iden-
tified 1 km northwest of Bayana/Sultankot in eastern Rajasthan dated from
the 1190s, or the period of the forays into northern India undertaken by the
Ghazna Sultan Mu‘izz al-Din and his military deputies (see Volume II).
The authors further claimed that an architectural precedent for this type
of prayer space could have existed at the eleventh-century South Palace at
Lashkari Bazar, specifically the so-called Great Mosque of the Forecourt.
However, notwithstanding the unusually elongated plan of this structure, it

180 Iran to India
is unlikely to have been an ‘idgah: remains of columns in both phases of the

building indicate that it was roofed (cf. Schlumberger 1978: 67–79 and pls. 23,
93–96). Nevertheless, it is of course possible that the simple, open-air prayer
area comprising an ‘idgah was a known architectural type in Khurasan, and
that examples pre-dating the fourteenth century both in Iran and India have
tended not to survive precisely due to their meager physical “footprints” – as
would have been the case at Jam-Firuzkuh. For a no longer extant Tughluq-
period (mid-fourteenth century) ‘idgah at Multan, cf. A. N. Khan 1983:
177.
34. In her analysis of architectural descriptions in the panegyric poetry (qasida)
produced at the Ghaznavid and Saljuq courts, Meisami (2001: 30, 41–42)
highlighted the growing abstraction in poetic references to palaces; over
time the poetry contained fewer descriptive details, increasingly substituted
with metaphors and rhetorical flourishes by the mid-twelfth century. By way
of explanation, she proposed both a broadening base of patronage beyond
royalty to the religious and military elites, and an attendant shift in the
very function of the palace from a forum of artistic creativity to a display
trophy. The less than illustrious origins of the Shansabanis not only called
for all the royal symbolism and legitimation of power that could be mus-
tered, their chronicler Juzjani was heir to the poetic-literary traditions of the
literati patronized by the Ghaznavids and Saljuqs. Understandably, then, his
architectural descriptions served largely symbolic rather than documentary
purposes.
35. Cf. esp. Flood 2002 for the historical and modern meanings of the Bamiyan
Buddhas’ destruction; for Bamiyan’s “posthumous” designation as a
UNESCO site, see Manhart 2006. See Inaba 2019 for the changing receptions
of Bamiyan’s Buddhas from the tenth through nineteenth centuries.
36. Cited by Tarzi 2007: 98. See also Thomas 2018: 58–59; Ball 2019: No. 100, for
a thorough bibliography.
37. Cf. de Planhol (1988); Klimburg-Salter 1989: 10–11; Ball 2008: 164–173; Lorain
2018. This conception of the Bamiyan basin also emerges from Baker and
Allchin’s (1991) excavations at Shahr-i Zuhak and related sites (see infra in
main text), and from Lee’s (2006) documentation of the historical fortifica-
tions in western Bamiyan Province. For Funduqistan, Fuladi, and Kakrak, see
also Ball 2019: Nos. 330, 332, 508, respectively.
38. Manhart 2006: 49; Tarzi 2007: 97, 107, 118ff.; Klimburg-Salter 2008: 132, 135,
140. Ball, Bordeaux et al. (2019: 391 and notes) espouse a date of the mid-
seventh century for the Buddhas, however.
39. Cf. Rezakhani 2017: 141–142, and ibid., 164–165 for a brief discussion of these
debates. See also Tarzi 2007: 121–122; Crossley 2019: 50–53.
40. Klimburg-Salter 2008: 151.
41. Ball (2020) reiterated Klimburg-Salter’s (1989: 134–136) earlier assertion.
For the major site of Qunduz (along with associated mounds in the vicin-
ity), see Ball 2019: Nos. 194, 569, 930, 931, 1160. Such a politico-dynastic
geography would be continuous with that of the Kushanas (first–fourth
centuries ce), whose political base was at Bagram but lavish patronage was
dedicated to their dynastic center of Surkh Khotal. Cf. also Ball 2019: Nos.

122, 1123, respectively; and Ball, Bordeaux et al. 2019: 350–351, 353–360,
376–379.
42. The tandem existence among these differing axes of communication between
Central Asia and India – the north–south axis via Bamiyan also incorporating
the more southerly reaches of Zabul and Zamindawar – nonetheless resur-
faced during the Ghaznavid and later periods. Cf. Inaba 2013: 90.
43. Perhaps an exception to this overall emphasis on Buddhist monastic and
religious architecture in the contiguous valleys was Kakrak, to the southeast
of Bamiyan: here, to the north of the grottoes, a small walled structure (fort?)
has been dated to the early phase of Shahr-i Zuhak, i.e., c. 500 ce. Cf. Le
Berre 1987: 81; Baker and Allchin 1991: 157. For a brief description of Kakrak’s
Buddhist ritual and monastic remains, see Tarzi, vol. I, 1977: 78. A descrip-
tion of the extremely rich grottoes of Fuladi was provided by Dagens 1964:
43–48, figs. and pls.
44. Cf. Le Berre et al. 1985–1988: 776, 782, 783; Lorain 2018. While Le Berre
and others have noted the topographical and architectural parallels between
the Bamiyan basin and Ghur – the latter being less clearly defined given
the lack of both historical and modern documentation there – they did
not pursue these observations further, to conclude that in Ghur much of
the architectural landscape was reused and repurposed by later populations.
Cf. Chapter 2.
45. The Bamiyan valley and specifically the Funduqistan monastery – the valley’s
only free-standing mud-brick structure (other complexes principally being
grottoes of various dimensions) – has offered evidence of Brahmanical–Hindu
cultic presence, similar to the coexistence of Buddhist and Brahmanical cults
at various sites in the vicinities of Kabul and Ghazna. Worthy of special
mention was the exquisite Mahisha-mardini from Funduqistan, preserved
only in Durga’s multiple arms and zoomorphic demon in marble (now lost).
Cf. Bernard and Grenet 1981; Klimburg-Salter 1989: 73–74; Ball 2008: 188–
189; Ball 2019: No. 332; and Chapter 2.
46. See Le Berre 1970: 45ff.; Baker and Allchin 1991: 55ff., 99–100; Lee 2006:
243–244; Thomas 2018: 118–119 (brief discussion); Ball 2019: No. 1052; Ball,
Bordeaux et al. 2019: 391. Fischer (1969: 344) and Baker and Allchin (1991:
99) proposed that in the Bamiyan structures larger bricks were generally older
than smaller ones, the latter likely indicating Islamic-period renovations or
consolidations of existing structures. However, brick sizes are notoriously
inconsistent, and only serve as a general indication of pre-Islamic or Islamic
period construction (W. Ball, pers. comm., April 2018).
47. See esp. Le Berre et al. 1985–1988: 780–781. Centuries-long occupation and
repeated renovations were evident at Shahr-i Zuhak itself where, during
the Timurid period (fifteenth century), apertures possibly for firearms were
created on the perimeter walls. Cf. Baker and Allchin 1991: 99. The virtually
uninterrupted use of ancient structures is not unique to the Bamiyan region,
also having been noted in Gandhara (Fischer 1969: 359) and widely docu-
mented in Sistan. Cf. Fischer 1969: 354; Fischer 1970: 484–485; Fischer 1973:
138, 143.
48. Cf. Baker and Allchin 1991: 35ff., 99. For Ghur, see Chapter 2.

182 Iran to India
49. Baker and Allchin 1991: 61.

50. Lee (2006) documented the architectural remains to the west of Bamiyan
proper; see also infra in main text, and Thomas 2018: app. 3.
51. For Sistan, cf. Fischer et al. 1974: illus. 36–38, 41, 43. Fischer (1971: 46,
50) also noted the regional developments within Sistan of certain building
types and, above all, techniques of decoration, particularly the combination
of mud-brick structures having a combination of stucco and burnt-brick
decoration.
52. Bamiyan, together with the rest of Tukharistan, Balkh, Gharjistan, and Ghur,
were given to Sabuktigin as a reward for his admirable victories in Khurasan
on behalf of his Samanid overlords; cf. Bosworth 1973: 44; Baker and Allchin
1991: 22–24. According to al-Muqqadasi (c. 985 ce; cited in ibid.), by the
time of Ghaznavid presence in the area in the tenth century, the town of
Bamiyan was walled and had a congregational mosque, thriving market,
several suburbs, and four city gates. Also noteworthy is the possibility that,
both during and after Sultan Mahmud’s reign and the Ghaznavids’ firm hold
over territories spanning from Ghazna through Balkh, wealthy non-Muslim
merchants were still prominent in Bamiyan: Baihaqi praised the merchant
Abaveyh, who rebuilt the bridge of Bamiyan “with a single arch, displaying
great elegance and beauty” (Baihaqi [trans. Bosworth] vol. I, 2011: 367). The
merchant’s name appears to be a Persian transliteration of the Indic (perhaps
Buddhist) Abhaya; although, Ashtiany offers the unlikely alternative that
the name was a hypocoristic form of ‘Abd Allah (ibid., vol. III, 2011: 167).
Moreover, the Shars as the pre-Islamic dynasts of Bamiyan were kept in place
during the Arab-‘Umayyad and A‘bbasid incursions, not being definitively
eradicated even after the Saffarids’ late ninth-century successes there (Le Berre
et al. 1985–1988: 782–783).
53. Juzjani vol. I, 1963: 348, 384 (trans. vol. I, 1881: 344, 361–362). Evidently,
Fakhr al-Din Mas‘ud appeared innocent of any collusion with the Madin
Shansabanis, thus retaining the Bamiyan region as his appanage, passed on to
his descendants for another three generations. By Juzjani implied that, by the
reign of Ghiyath al-Din Muhammad (ah 553–599/1163–1203 ce) at Firuzkuh,
the paramountcy of the Firuzkuh Shansabanis was firmly established over the
others, as indicated by Ghiyath al-Din’s personal conferral of the title “sultan”
upon his cousin Shams al-Din Muhammad (r. ah 553–588/1163–1192 ce),
perhaps the first of the Bamiyan rulers to receive it. See Juzjani vol. I, 1963:
378 (trans. vol. I, 1881: 427); see also infra in main text and notes; and for the
contentious relations among the Shansabani branches, see Chapter 1.
54. Cf. Juzjani vol. I, 1963: 388 (trans. vol. I, 1881: 428–430); and Chapter 2.
55. While in the service of the Saljuq sultans Muhammad I (r. ah 498–512/
1105–1118 ce) and Sanjar (r. ah 512–552/1118–1157 ce), Nizami ‘Aruzi likely
moved with the court throughout Khurasan, as glimpsed in the autobio-
graphical introduction of the Chahar Maqala (cf. also Peacock 2015: 167).
The circumstances in which he joined the Firuzkuh Shansabani court remain
unknown, though it would be intriguing indeed to have his eyewitness view
of his erstwhile patron Sanjar’s defeat and capture of ‘Ala’ al-Din Husain near
Herat in 1152/3 ce, and the reasons for Nizami ‘Aruzi’s own stay in hiding

in the city afterward. See Gholam-Hosayni 1990; also Nizami ‘Aruzi (trans.
Browne 1921) 1978: 7–11.
56. Cf. Ghafur 1960: 2, 150–151; Anawati 2012.
57. Juzjani’s dedication of an entirely separate section (XVIII) of Tabaqat to the
“Shansabaniyya of Tukharistan and Bamiyan” contains the lengthy passage
on his father Maulana Siraj-i Minhaj’s move to Bamiyan from Firuzkuh.
Historiographically, such an organizational device in Tabaqat places the
Bamiyan branch of the family on a par with the one at Jam-Firuzkuh. It
could also be that, given his encomium of Maulana, Juzjani separated and
highlighted the Bamiyan branch as much to attest to its status, as to pay
homage to his father’s erudition and piety. Cf. Juzjani vol. I, 1963: 389 (trans.
vol. I, 1881: 430); also Nizami 1983: 76–77. Edwards (2015: 112) cited another
example of Juzjani’s subordination of historical events to individual narrative
in his all too brief report on the Ghazna Shansabani expansion into the Indus
Valley (Chapter 6). Rasikh (2020: 113) described the emphasis of ties to well-
known personages as “a literary trope for medieval Muslim historians with
which they document consciously their personal, familial, and professional
biographies.”
58. Cf. Juzjani vol. I, 1963: 389 (trans. vol. I, 1881: 429–431) with emphasis added.
59. It is well known that many areas lying in the Mongols’ eastward progress suf-
fered destruction, e.g., the Marv oasis where Mongol forces ransacked Sultan
Sanjar’s mausoleum while seeking precious objects as loot. See Hillenbrand
2011: 280; Jackson 2017: 175. Bamiyan was especially unfortunate, however,
to have been the place where Chinggis Khan’s favorite grandson Möe’tügen
was killed, for which the entire population was massacred and the town and
its surrounding valleys gutted – a level of ruination apparently still visible
through the fourteenth century. Cf. Gardin 1957: 228; and esp. Jackson 2017:
158. Indeed, Inaba (2019: n. 35) noted the absence of Bamiyan-minted coinage
after the thirteenth century – a possible indication that Mongol destruc-
tion had enduring politico-economic effects on the region. In contrast to
Bamiyan, it appears that only one Mongol incursion entered into the heart of
Ghur, heading southward from Kashi or Ahangaran and reaching the head-
waters of the Rud-i Ghur; while it is possible that the party pushed through
to Tiginabad/Old Qandahar, the regions of al-Rukhkhaj and Zamindawar
had always been more readily accessible via Ghazna from the east. Cf. O’Neal
2017; see also Chapter 2.
60. Gardin’s (1957) study of ceramics from the Bamiyan basin focused almost
exclusively on the finds from Shahr-i Ghulghula (see infra in main text),
which he contextualized within the ceramic production of the wider Islamic-
Persianate cultural sphere of Iran and Central Asia. It is only the cursory
but still very valuable study of 250 ceramic shards collected by Marc Le
Berre during his 1974–1975 surveys of parts of the Hindu Kush – principally
Bamiyan’s secondary valleys – that provides an overview of developments and
tentative ceramic as well as architectural chronologies of the region. Cf. Le
Berre et al. 1987; Le Berre et al. 1985–1988.
61. The work of Le Berre and his contemporaries of the Délégation Archéologique
Française en Afghanistan (DAFA) (1987; 1985–1988), and that of Lee (2006)

184 Iran to India
has already been cited for the Hindu Kush fortifications; for additional bibli-
ography see also Ball 2019: Nos. 189, 239, 286, 398, 489, 845, 862, 1021, 1039,
2045, 2046, 2141.
62. See Lee (2006: 229 and passim) for a “Ghurid” dating of the ruins, and Baker
and Allchin (1991: esp. 99) for the centuries-long use of decorative iconogra-
phies. The mural paintings from the western Bamiyan area were not unique
of course, but conceivably an extension of the lavish painted programs in
numerous grottoes and structures in Bamiyan and throughout the region’s
other principal valleys, particularly abundant in Funduqistan, Foladi, and
Kakrak. Indeed, the common curved striations in these paintings (Figure 3.18)
immediately evoke the nimbuses frequently seen behind the Buddha’s head:
e.g., Klimburg-Salter 2008: pl. V, det. 3. Cf. also Klimburg-Salter 1989: e.g.,
pls. XVIII, XXIII, XXIV fig. 26, LXXX fig. 103, and LXXXIV–LXXXVI. The
eighth–ninth-century caves in Jaghuri (west of Ghazna) likely also had some
paintings, though their unfinished (at Tapa Zaytun) or extremely eroded
states make it difficult to identify traces. Cf. Verardi and Paparatti 2004: 104
and pl. XLVIIa–c. See Chapter 2 for Herberg’s (e.g., 1982) documentation
of traces of interior mural painting within Ghur’s architectural remains. Pre-
Islamic architectural cultures throughout the vast central span of Afghanistan,
then, included painted programs.
63. Lee (2006: 238–241) also mentioned painted scenes from Shah-nama docu-
mented in 2002 at Chehel Burj (one of the farther western complexes, also
visited by Maricq and Wiet [1959]); unfortunately, the paintings disappeared
shortly afterward, probably looted and offered for sale at clandestine markets
in Pakistan or Iran. See also Thomas 2018: app. 3; Ball 2019: No. 189; Ball,
Bordeaux et al. 2019: 399.
64. Le Berre 1970: 45 (my trans.); also Fischer 1978: 351–352; W. Ball, pers.
comm., April 2018; Ball 2019: No. 109; Ball 2020; Ball, however, considers
Danestama’s “interpretation as a madrasa [i.e., newly built] … a convincing
one.” The mosque known as Darra-i-Sheikh at Gorzivan (Faryab Province)
presents a parallel to Danestama: also quadrangular in plan with circular corner
towers, and a more complete mihrab with stucco revetment and remains of
painted decoration. The additional surviving fragments of a floriated Kufic
inscription of Darra-i-Sheikh led to its tentative dating in the later twelfth
century, based on comparisons with inscriptions from Sar-i Pul (historical
Anbir) and the minar of Jam-Firuzkuh. Cf. Bivar 1966; esp. Pinder-Wilson
1980: 97–98; Ball and Fischer 2019: 487; Ball 2019: No. 248; and Chapter 2.
However, it is conceivable that Darra-i-Sheikh’s mihrab was also a later addi-
tion to the structure, as proposed here for Danestama (see following in main
text).
65. Le Berre (1970: 50) himself confessed to the collection of “only those ceramic
fragments that appeared most interesting to us and unpainted pottery was not
retained” (my trans.). Baker and Allchin’s (1991: 195–196) subsequent explora-
tion of Danestama resulted in the identification of pottery fragments and the
extant vaulting as comparable to those from Shahr-i Zuhak, c. 500 ce. A prime
example of the four-ivan Central Asian Buddhist monastery would be Ajina
Tepe (cf. Litvinsky [1984] 2011), and within Afghanistan the recently excavated

Mes Aynak provides a comparandum. See infra in main text for possible mili-
tary uses of the four-iwan structure with corner bastions, e.g., at Sarkhushak.
66. It is noteworthy that, as early as the eleventh century, Persian poets of the
eastern Islamic world such as Gurgani (fl. c. 1050 ce) were familiar with
Buddhist monasteries, whose square plans, central courtyards flanked by ivan-
like alcoves, and even interior paintings inspired metaphorical references,
or may have served as devices of mis-en-scène (cf. Melikian-Chirvani 1982:
6–7; and Introduction). This apparent first-hand familiarity with Buddhist
monastic interiors could have come from remnants of these structures in the
landscape, and/or the result of their not infrequent repurposing as madrasas
or other identifiably Islamic purposes.
67. Cf. Le Berre 1987: 88–93; Baker and Allchin 1991: 161–179; Thomas 2018:
118–119; Ball 2019: No. 1004. Baker and Allchin confirmed the dating with
analysis of surface ceramic finds, consisting mostly of sgraffiato wares and one
sherd of lusterware from Rayy, all datable to the late twelfth–early thirteenth
centuries.
68. It is indeed unfortunate that Juzjani did not describe Sarkhushak’s supposed
palatial structures in his section on the Bamiyan Shansabanis; they would
have surely lent themselves with ease to the metaphorical apprehension of the
palace as a symbol of royal greatness, common in eastern Persianate literary
culture during the twelfth century (cf. supra the section on Firuzkuh).
69. Quite possibly, Baker and Allchin’s Building E (Figure 3.25), occupying the
promontory’s highest point (to the north of the multi-level, graded resi-
dential/palatial structure), was a commemorative structure with a mihrab.
While the other mosque (Building F) exhibited an east–west rectangular plan
axially oriented along the qibla, Building E’s compact footprint and profile
essentially comprised a domed cube with addorsed entrance on the east – on
axis with the western wall’s mihrab – making it directly comparable to the
eleventh- or early twelfth-century ziyarat of Imam-i Khurd near Sar-i Pul
(ancient Anbir). Cf. Bivar 1966: esp. 57–62; Blair 1992: 201; Ball and Fischer
2019: 487.
70. The basic plan of the quadrangular structure (Building D), with its four
circular bastions at the corners, is clearly evocative of the possibly repurposed
Buddhist monastery of Danestama and other Central Asian Buddhist monas-
tic structures (see supra in main text). It should also be mentioned that dyed
and folded textiles were recovered from Building D, which of course does
not necessarily preclude its simultaneous function as a military garrison. Cf.
Baker and Allchin 1991: 163.
71. For the Indic elements cf. esp. Le Berre 1987: 92; Baker and Allchin 1991: 63,
65, 163. The introduction of Indic construction methods urges consideration
of a late twelfth-century date for at least some Sarkhushak structures, as the
Shansabanis’ Indian campaigns arguably brought more skilled laborers west-
ward from India, as proposed in Volume II.
72. The résumé here of the findings from Shahr-i Ghulghula emerged from the
invaluable archaeological fieldwork in 2014–2015 around the Bamiyan region
conducted by Thomas Lorain (Scientific Secretary, DAFA); while areas of
Shahr-i Ghulghula’s western slope were surveyed and some also excavated,

186 Iran to India
the topmost “citadel” still awaits exploration in upcoming seasons. In the

2014–2015 tours, Lorain and his team also encompassed the site of Chehel
Dukhtaran to the south of Shahr-i Ghulghula. Cf. Lorain 2018; also Ball 2019:
No. 1042.
73. See esp. Gardin 1957; cf. also Gascoigne 2010 for discussion of parallels and
divergences between ceramic finds at Shahr-i Ghulghula and Jam-Firuzkuh.
Now that the ceramic collections of the Herat Museum have undergone
analysis, further parallels will surely emerge and better contextualize the
Bamiyan–Ghulghula assemblages: cf. Franke 2016b, 2016c; Müller-Wiener
2016; Franke and Müller-Wiener 2016. Ultimately, all of these considerations
must be framed against the glaring need to revisit the entirety of Gardin’s
collected ceramic corpus and his conclusions, not least given the fact that the
materials were largely gathered through purchases at local markets rather than
from documented archaeological contexts; and the sheer quantity of remains
that would have probably exceeded the very short period of production pro-
posed by Gardin (T. Lorain, pers. comm., September 2018).
74. Cf. Godard 1951: 5, figs. 1 and 4; Ball 2019: No. 1042.
75. See Baker and Allchin 1991: 28–30; Rugiadi 2006; Ball 2008: 166, 171; Thomas
2018: 217, 220. Lorain (pers. comm., September 2018) noted that the wooden
doors’ whereabouts now are unknown. While Warwick Ball (pers. comm.,
April 2018) confirmed Shahr-i Ghulghula’s pre-Islamic construction – citing
a Buddhist stupa and a hoard of T’ang Chinese bronzes recently documented
there – he also noted a fundamental difference between Zuhak and Ghulghula:
while Ghulghula was “the more urban site for Bamiyan (albeit fortified) …
in use down to the Mongol conquest, [Zuhak was] solely a fortification.”
Ghulghula’s high-quality ceramics, the intricately carved wooden doors, and
the recent documentation of probably Ghaznavid-period palatial remains
(cf. below in main text) would suggest multiple functions for the complex,
including a royal residence/palace. Cf. Godard 1951; Gardin 1957; Baker and
Allchin 1991: 28–30.
76. Cf. also Godard 1951.
77. R. Giunta, pers. comm., July 2018; Giunta 2020; cf. also Chapter 5.
78. Lorain 2018.
79. Cf. Bosworth 1988; Baker and Allchin 1991: esp. 163; and see Chapter 4. See
also O’Neal (2020: 205–207) for numismatic evidence elucidating the nature
of the Shansabanis’ periodic occupations of Balkh as they confronted their
various adversaries, including the Khwarazm-shahs in the 1190s ce.
80. Danestama is comparable with another, somewhat mysterious and probably
multi-purpose structure northeast of Multan (Pakistan), just east of the Indus
River’s current flow. It was newly constructed in the third or fourth quarter
of the twelfth century, though documentation by the present author in 1998
took note of older fragments used as building material in the walls; moreover,
the Indus structure’s plan very much echoes Danestama (Figures 6.16 and
6.17). Cf. Dani 1982: 183; Mumtaz 1985: 39–42; Edwards 1991; and Chapter 5.
81. See esp. Inaba (2013), and the discussion of the many Ghaznavid palaces (e.g.,
Balkh, Ghazna, Bust-Lashkari Bazar, Nishapur), palace–garden complexes
(e.g., also Nishapur, Herat, Bust), as well as royal tent complexes used by the

Yamini sultans, frequently characterized as “Iranized Turks” (ibid., 75) due

to their extensive movements across far-flung tracts and occasional residence
within “textile architecture.” The sultans’ movements also combined seasonal
transhumance with military expeditions (ibid., 89).
82. The modern city (Shahr-i Nau) swallowed within it the quadrangular, walled
city laid out by Ahmad Shah Durrani c.1755, itself replacing Nadirabad, the
garrison established by the warlord Nadir Shah Afshar (d. 1747) in the 1730s
as he was on his way to plunder the Mughal treasury in Shahjahanabad (Old
Delhi). Cf. N. Dupree 1977: 279ff.; Matthee and Mashita [2010] 2012); and
infra in notes.
83. Symptomatic of the multiple and changing names of Old Qandahar is its
sobriquet “Shahr-i Kuhna,” which is also applied to a locality outside the city
of Farah (Farah Province), and to another site in the region of Zamindawar
(Helmand Province): cf. Ball 2008: 228–233; Ball 2019: Nos. 522, 1047; and
Chapter 2. For the time frame of the present study, Ball’s (1988) observa-
tions are important to bear in mind: the Old Persian Harahuvatish for the
entire region was Hellenized as Arachosia, and in turn Arabicized as al-Rukh-
khaj in early geographies (ninth–tenth centuries). Likely due to the Turkic
Ghaznavids’ presence in the area, the locality came to be called Tiginabad
as of the eleventh century. It was not until the thirteenth century that the
name “Qandahar” was applied to the city. Cf. Ball 1996: 400–401; Ball, pers.
comm., February 2018; Bosworth 2012b.
84. For the presence of an eighth-century ‘Umayyad coin hoard in the Qaitul
vihara, see Helms (1982: 14), who further noted the possible “spiritual coex-
istence” of the local Zun/Zhun cult plausibly dedicated to a solar deity,
Buddhism and Islam. See also Helms 1983; Ball 1996: 399; Errington 2017:
43; and Chapter 2 notes. The extensive documentation and analysis of the
Qaitul Buddhist complex unfortunately still awaits publication, even though
the British Institute for Afghan Studies conducted surveys and excavations
throughout the entire area – encompassing both the Old Qandahar citadel
and the stupa overlooking it – in their 1974–1978 seasons (Ball, pers. comm.,
February 2018).
85. Ball (1988: 135–137) proposed that, given the appearance of the locale’s name
as “Qandahar” only in sources dating from the thirteenth century and later,
the Shansabanis could have initiated the name change: the association of
“Tiginabad” with the recognizably Turkic presence of their enemies the
Ghaznavids surely spurred the Shansabanis to appropriate the site by means
of naming it with a term rooted in discernibly local Persianate traditions,
both as a sign of victory and of their own induction into a prominence
outside their ancestral Ghur. But the fact that Juzjani referred to the loca-
tion only as Tiginabad, making no mention whatsoever of Qandahar, would
indicate that the name change did not originate with the Shansabanis, but
probably later than the sparse traces of their presence there (see infra in
main text).
86. Cf. Juzjani vol. I, 1963: 243 (trans. vol. I, 1881: 111). Nevertheless, the debate
surrounding whether Qandahar and Tiginabad are one and the same needs
“considerable … work … before [it] can be put to bed”: Ball (pers. comm.,

188 Iran to India
April 2020) has put forth the additional possibility of Kushk-i Nakhud, a.k.a.
Sabz Qala‘ (66 km west of Qandahar) as Tiginabad, as the site might have
been larger than Qandahar itself, and ceramic finds there have been dated
from the first through fifteenth centuries ce.
87. Ball 1988: 138 (original emphasis), and Ball 2008: 230; see also Helms 1982:
15; Helms 1997: 3–4. The city reclaimed real imperial significance only when
the Timurid ruler Husain Bayqara (r. ah 874–912/1470–1506 ce) established
a mint there in the late fifteenth century. Cf. Inaba 2010.
88. The depredations of Nadir Shah Afshar (1688–1747) and his forces during
his India campaign – where the aim was plunder of the Mughal treasury
and libraries – resulted in considerable destruction in the vicinity of Old
Qandahar in the mid-1730s. This in turn has rendered a full reconstruction of
earlier phases of the city quite difficult. Cf. Whitehouse 1976: 473; Ball 1996:
402. The Afsharid soldier-adventurer’s initial choice of the southern route via
Qandahar in 1736 was clearly strategic, rather than commercial or practical,
however: after their brief sojourn in the Qandahar region, the Afsharid forces
ultimately turned northeast toward Kabul and Jalalabad. They entered India
via the Khyber Pass in 1738–1739, visiting intimidation and destruction upon
all in their path, whenever necessary. See Dupree 2012: 330–331.
89. Ball (1988: 135) suggested that the Shansabani presence in Tiginabad/Old
Qandahar commenced in ah 521/1127 ce, which would fall within the reign of
‘Izz al-Din Husain (r. 1100–1146 ce). The nature of the archaeological evidence
(see infra in main text) nonetheless points to the later date of c. ah 552/1158 ce,
when the Shansabanis were a more important politico-military and economic
presence in the region. Moreover, the confusion between ‘Izz al-Din and
‘Ala’ al-Din Jahan-suz in Khwafi’s fourteenth-century Majmu‘al-i Fasihi – the
source noting ah 521/1127 ce for the Ghaznavid–Shansabani confrontation
– was actually cited by Raverty in his copious footnotes to Juzjani’s text (cf.
trans. vol. I, 1881: 110, n. 5). Again, the later date as noted by Juzjani himself
seems the more probable. The at best indirect references to these events in two
contemporaneous sources – viz. Juzjani and Fakhr-i Mudabbir – have added
to the tentativeness of dating. Cf. also Bosworth [1977] 1992: 121–122.
90. See also Chapters 1 and 5. Sultan Ghiyath al-Din granted the fort of Istiya to
his as yet untitled but ambitious younger brother Shihab al-Din, likely in the
mid-1160s (cf. Chapters 4–5). During the previous generation of Shansabanis,
the region had been the appanage of Saif al-Din (d. ah 544/1149 ce) who,
among ‘Izz al-Din’s seven sons was the third born, but the first by his lawful
wife and thus took precedent over his two older half-brothers. The locality of
Istiya can be tentatively placed within the Shansabanis’ garm-sir. Cf. Juzjani
vol. I, 1963: 335–336 (trans. vol. I, 1881: 339–340).
91. See Whitehouse 1976: 484; Whitehouse 1996: 4–5; Ball 1996: 400; Ball pers.
comm., February 2018.
92. Two excavation reports by Whitehouse (1976 and 1996) reflect different
stages of the archaeological work. For example, the earlier report (1976) was
published prior to the rediscovery of the second burial in Mound A.
93. Cf. Whitehouse 1976: 483; Whitehouse 1996: 4; Crowe 1996: 314; Ball,
Bordeaux et al. 2019: 433–434. Whitehouse in particular proposed parallels

with Lashkari Bazar Group IX (see Gardin 1963: 135). Additionally, Taddei
(1979: 910ff.) suggested that the similarity between the octagonal enclo-
sure-platform of Mound A and the tomb known as Sultan Ghari in Delhi,
dated to the 1230s ce, lend formal and circumstantial evidence for Shansabani
dating and patronage of Tiginabad/Old Qandahar’s Mound A (and possibly
Mound B). The parallel with the later Sultan Ghari is also mentioned by
Edwards (2015: 102).
94. Hephthalite burials in Afghanistan certainly provide comparable parallels
to the excavated Tiginabad/Old Qandahar mounds, and with good reason,
given the Rutbil command of the area well into the ninth century. It is
also noteworthy that mound burials, or kurghans, were quite different from
urban burial practices and have been “associated with the nomadic nations of
Central Asia”; some scholars have used the dissemination of kurghans to trace
movements of nomadic groups. Cf. Ball 1996: 400–401. But, cf. Hüttel and
Erdenebat (2010: 4–5) on the study of Eurasian nomadic horsemen since the
seventeenth century as essentially a one-sided “archaeology of graves … a cul-
tural history … limited to the history of burial rituals …” desperately needing
to be balanced with the archaeology of nomadic settlements for a “more
authentic picture of the cultural scope and diversity of nomadic empires of
the steppe …” See Chapter 5; and Volume II for the more orthodox Islamic
burial and tomb of the later patriarch Sultan Ghiyath al-Din Muhammad
(d. ah 599/1203 ce) in Herat.
95. In the context of the events unfolding in the 1160s–1170s, Juzjani vol. I,
1963: 396 (trans. vol. I, 1881: 448–449) had described Tiginabad as the largest
city in garm-sir, as well as the pivot of the Ghaznavids’ downfall when the
Shansabanis captured it. Quite possibly, the Oghüz bands fleeing Saljuq
reprisals in Khurasan only accelerated the gradual depopulation of the region
(see supra in main text). Cf. also Bosworth [1977] 1992: 124–125. Thus, even
Tiginabad might have appeared to be a large city in a region largely emptied
of its inhabitants. Tiginabad could also be seen as the origin of the Yaminis’
irreversible defeat, since it was from there that the Shansabanis launched
successful operations against Ghazna itself.
96. Cf. Thomas 2018: 135–136; see also supra in main text and notes.
97. Cf. de Planhol [2000] 2012.
98. The family hierarchy was not without contestation, as is evident in the
Bamiyan Shansabanis’ adoption of royal titles on their coinage several decades
prior to their official conferral – see Chapter 1.
99. Hüttel and Erdenebat 2010: 5.

CHAPTER 4
One and Several: Gharjistan, Chisht,

and Imperial Firuzkuh
T he architectural corpus analyzed here spans approximately the late

1160s–1170s, encompassing the Firuzkuh Shansabanis’ sard-sir
“capital” and other sites radiating westward toward Herat, and northwest
to Gharjistan. The buildings and complexes consist of the following exam-
ples, discussed in the proposed chronological order, which is, admittedly,
at some variance with previous scholarship: The madrasa known as Shah-i
Mashhad near Jawand in Gharjistan (Figure 4.1), which was founded in
ah 561/1164–1165 ce and likely completed a decade later, probably also
included a tomb (Figure 4.5).1 Chisht’s two domed structures, one with
a mihrab (Figures 4.12 and 4.17), evidently belonged to a larger complex
that was possibly a madrasa, located east of Herat and dated by inscrip-
tion to ah 562/1167 ce (construction and renovation probably extending
before and after this date). Finally, the famous minar of Jam-Firuzkuh
(Figure 4.18), also discussed in Chapter 3, has an inscription that was
re-read as dating to ah 569/1174 ce, superseding the previous reading of
ah 590/1193–1194 ce (cf. below). All of these buildings and sites are impor-
tant surviving examples of monumental architectural initiatives within the
purview of the Firuzkuh Shansabanis.
Although the Bamiyan Shansabanis – discussed in Chapter 3 – doubtless
undertook architectural projects in their territories, the absence of firmly
datable evidence makes further examination of their patronage quite diffi-
cult, with the possible exception of forays into the Balkh oasis.2 It should
also be noted that, by this period, the Firuzkuh and Ghazna Shansabanis’
building activities included, respectively, monumental architecture and
extensive interventions in pre-existing complexes, the latter phenomenon
also calling for a concentrated analysis in its own right in Chapter 5. As
will be subsequently discussed in detail, the documentable architectural
footprints specifically of the Ghazna Shansabanis in Afghanistan – and by
extension those of the architectural culture of the eastern Persianate world –
were particularly impactful throughout their campaigns farther east, first to
the Indus’s alluvia and ultimately to India (see also Volume II).
A significant difference between the preceding chapter’s earlier archi-
tectural group and the one analyzed here is the extensive architectural
epigraphy on the latter, at times warranting discussion at some length (cf.

One and Several: Gharjistan, Chisht, and Imperial Firuzkuh 191
later in this chapter and Appendix). As we shall see below, the epigraphic
programs of these later foundations frequently reveal theological debates,
shifts in legal thought, and far-reaching changes in the nexus of religion
and politics, all occurring within the expanding realms of the increas-
ingly divergent Shansabani lineages of Firuzkuh and Ghazna – one ori-
ented westward toward Khurasan, the other eastward toward India (see
Chapters 5 and 6). Altogether, prior studies of the Gharjistan and Chisht
madrasas, the previous and recent analyses of the minar of Jam-Firuzkuh,
as well as excavations at the site, provide invaluable points of depar-
ture for better understanding the impetuses and processes behind the
making of these buildings and complexes, especially as they became sacral
institutions.3
Architectural patronage certainly evidenced the Shansabanis’ active
presence in an area, perhaps indicated earliest at Da Qadi/Qala‘-yi
Zarmurgh and the site’s minar, likely dating to the first two or three
decades of the twelfth century (see Chapter 2). But as of the 1170s onward,
Shansabani patronage unequivocally demonstrated the mobilization of
a different caliber of skilled labor: with Shansabani foundations reflect-
ing the cosmopolitan building trends prevalent in elite circles across
the Persianate world, it is evident that their resources were rapidly bur-
geoning beyond those of a few decades before. Finally, it would appear
that, while the Firuzkuh Shansabanis retained a lifestyle of seasonal
transhumance, the Bamiyan and Ghazna branches constituting the rest
of the confederacy might have been largely shifting toward a sedentism
marked by palace dwelling, at least for part of the year (see Chapters 2, 3,
and 5).
Shansabani Imperial Architectures
Existing studies on Shansabani-patronized buildings and complexes in

Afghanistan have approached them collectively, characterizing them overall
as “a remarkable range of civil, religious and commemorative architecture
… [that was] part of a coordinated campaign to champion Islam”.4 But
we saw in Chapter 3 that, in fact, inescapable differences emerged among
the indices of Shansabani presences at early Firuzkuh, Bamiyan, and
Tiginabad/Old Qandahar, at least according to the surviving evidence.
The early Shansabanis at Firuzkuh appear to have eschewed patronage of
palatial architecture, despite its being de rigueur for Perso-Islamic kings
according to their chroniclers and panegyrists. Instead, they concentrated
on more “civic” foundations such as places of prayer, and probably defen-
sive structures. Tiginabad/Old Qandahar was a site likely occupied by the
Firuzkuh Shansabanis, a territory morphing into an expansionist political
entity; but excavations and surveys there unearthed fewer architectural

192 Iran to India
traces of Shansabani presence, the more nomadic traces not being easily
recoverable. On the other hand, the Bamiyan Shansabanis’ architectural
patronage seemed to be focused largely on repurposing pre-existing struc-
tures and complexes. This was evidently the case at the numerous forti-
fications and citadels to the west of Bamiyan, including Shahr-i Zuhak.
Furthermore, at Shahr-i Ghulghula Ghaznavid-period structures were
apparently put to Shansabani use. Although new construction in the
Bamiyan region occurred less frequently, the newly built palatial complex
at Sarkhushak nonetheless indicated a court life and ceremonial gradually
diverging from that of their Firuzkuh cousins.
The differences between the earlier commissions discussed in Chapter
3, and the monumental foundations at later Firuzkuh and its ambit, as well
as the interventions at Ghazna and Bust-Lashkari Bazar (Chapter 5), are
even more significant. Thus, the juxtaposition of all of these architectural
corpora – such as they are known from their surviving traces – makes it
clear that any singular, overarching characterization such as “Shansabani/
Ghurid architecture” would be untenable.
Here, it is important to note that what might appear to be a largely
silent, background process must in fact be foregrounded in the study of
an empire emerging from transhumance and/or nomadism and consisting
of various lineages: given the particularly complex and often piecemeal
process of Islamization among nomadic and many other non-elite groups
(cf. Chapter 2), it is necessary to re-examine the scholarly assumption that,
in general, patrons determined the ultimate outcomes of their projects.
Actually, how much interest did the historically and even persistently
transhumant Shansabani elites of Firuzkuh and Ghazna – though the
latter increasingly engaged with already existing palace architecture (see
Chapter 5) – have in determining the architectural forms and epigraphic
programs of the buildings they commissioned? What did it fundamentally
mean for nomadic, seasonally transhumant, or gradually sedentizing elites
to patronize monumental architecture? Addressing these questions helps
to shed light on the nature, inner workings, accumulation, and exer-
cise of power by the Firuzkuh Shansabanis. In the context of historically
transhumant/nomadic elites crafting a loosely intertwined imperium –
and one spanning multiple cultural geographies – we must revisit the role
and input of the patron in the design of buildings and their inscriptions,
as well as the reception of the structures and their epigraphic programs by
the many groups using and maintaining them.
By means of close examination, the Firuzkuh Shansabanis’ large-scale
architectural projects elucidate the encounter between more nomadic
and sedentizing elites and the Perso-Islamic cultural complex. The latter
encompassed – but was not limited to – Persianate literary and other
cultural productions as well as what might be termed for convenience
the “Great Tradition” of Islamic orthodoxies (cf. Chapter 1). The internal

divisions of these orthodoxies and the resulting, periodically violent con-

tests for intellectual, juridical, and politico-economic sway, were especially
legible in the epigraphic programs on monumental architecture: specific
Qur’anic inscriptions and Hadiths came to be appropriated by some
madhhabs to express their distinctive theologies.5 As a result, some archi-
tectural foundations also proclaimed their sectarian affiliations through
their epigraphic programs, as was the case for the Jam-Firuzkuh minar.6
The ensuing examination traces the dialogue between the architecture of
the Firuzkuh Shansabanis and the specific cultural geography they inhab-
ited, which was one among several cultural–geographical zones in the
region now bounded by the nation-state of Afghanistan (cf. Chapter 1).
Ultimately, an apprehension of the role of architectural patronage in con-
structing an empire – literally and figuratively – is important not only
for elites emerging from transhumance/nomadism, but for all historical
state-builders.
The ensuing chapter encompasses Shansabani architectural patronage
up to the definitive conquest of Ghazna in ah 568–569/1173–1174 ce, to
which date the famous minar of Jam-Firuzkuh has been convincingly
reassigned (see below). The aggregation of the corresponding architec-
tural corpora is justifiable not only due to this significant historical event,
but also by the locations and characteristics of the surviving buildings:
the Chisht and Gharjistan complexes and the Jam-Firuzkuh minar were
new, monumental foundations located within the broader architectural
culture of Khurasan – at the eastern edges of the Persianate world – with
substantial and extremely meaningful epigraphic programs evincing their
localities’ palpable participation in the theological and political develop-
ments throughout the eastern Islamic lands. Meanwhile, the Shansabani
vestiges at Ghazna and Bust-Lashkari Bazar, in the Zabul–Zamindawar
corridor (map, p. xvi), were altogether of a different order, being primar-
ily interventions in the architectural fabric of Ghaznavid-period palace
complexes, rather than new constructions per se (see Chapter 5). The dis-
tinguishable and localized historical geographies of these contiguous areas,
and their overall significance for the Shansabanis as a confederacy, are best
understood by means of clustering and juxtaposing these architectural
groups.
Gharjistan, Chisht, and Imperial Firuzkuh
As is to be expected after our previous searches for even brief, contem-

porary textual descriptions of Shansabani-patronized architecture (cf.
Chapters 2 and 3), little information is forthcoming in written sources.
The lengthy section XVII of Juzjani’s Tabaqat, with many chapters on
the reigns and activities of the Shansabani maliks and sultans of Ghur and

194 Iran to India
presumably one of the principal foci of this particular work more broadly
(see Chapter 1), nevertheless mentioned very few specifics regarding the
locales where the Firuzkuh Shansabanis resided or brought under their
sway. To wit, the continuing search for the winter “capital” of Zamindawar
was discussed in Chapter 2, while the descriptive vagaries pertaining to
Firuzkuh itself, and the resulting debates surrounding its identification as
the Shansabani summer “capital,” have been addressed in Chapter 3 and
elsewhere.
Juzjani also made no mention of Chisht, about 150 km east of Herat,
where two domed structures are what remain of probably another impor-
tant complex, which originally had an extensive program of cursive and
Kufic inscriptions, three of which referred to Shansabani rule (Figures
4.12–4.15, 4.17).7 Remnants of the cursive inscription on the interior of the
better preserved southern structure (Figure 4.12 [left structure], 4.13, and
4.14), ringing the base of its dome, named Shams al-Dunya wa al-Din,
presumably a reference to Sultan Ghiyath al-Din, “in [whose] days the
renovation (tajdid) of this building was ordered” (Figure 4.14); the Kufic
inscription ending on the east interior wall (Figure 4.15) lists the date of
10 Jumada al-Awwal 562 ah, or 7 March 1167 ce, which, it should be
noted, would render anachronistic the use of the sultan’s non-regnal laqab
(discussed below).8
One partial exception to Tabaqat’s overall laxity regarding locations
and their descriptions (cf. Chapter 2) was the region of Gharjistan, north-
west of historical Ghur, which entered Juzjani’s narrative multiple times:
according to the author, shortly after Qutb al-Din’s departure for Ghazna,
his brother Baha’ al-Din (d. 1149) struck an alliance with the Shars of
Gharjistan during the early days of Shansabani ascendancy at Firuzkuh,
allowing the reinforcement of Ghur’s defenses via the construction of new
fortifications there and elsewhere. After Baha’ al-Din’s death, ‘Ala’ al-Din
Husain Jahan-suz called up forces from Gharjistan in ah 545/1150 ce as he
prepared for his momentous confrontation with Bahram Shah of Ghazna
on the plains of Zamindawar (cf. Chapter 1). ‘Ala’ al-Din Husain further
cemented the Firuzkuh–Gharjistan alliance by marrying Hurr Malika, the
daughter of Shar Ibrahim ibn Ardashir.9 It would appear that Shansabani
presence in Gharjistan – and quite possibly the alliance between the two
ruling houses – continued through the 1160s–1170s. At this time, there was
a madrasa at Afshin – an unmapped locality, which Juzjani incidentally
noted as the capital of the Gharjistan Shars.10 The madrasa’s importance
lay in its presiding head being the prominent Karrami Imam Sadr al-Din
Ali Haisam of Nishapur.11
Also dating to this period is the foundation inscription of Shah-i
Mashhad, which unequivocally indicates that a female patron – presumably
Taj al-Harir Jauhar Malik, wife of Sultan Ghiyath al-Din – commissioned
the magnificent complex (Figures 4.1 and 4.7).12 While Shah-i Mashhad’s

function as a madrasa has long been accepted, recent re-examination of

its remains and the elaborate inscriptional program led A. W. Najimi13
to propose that the complex’s southeast corner contained the tomb of
Saif al-Din (r. 1161–1163) (Figure 4.5), ‘Ala’ al-Din Jahan-suz’s son. Saif
al-Din was betrayed and treacherously assassinated in ah 558/1163 ce by his
sipahsalar ‘Abbas ibn Shish, who belonged to the long-time rival Shishani
clan,14 while the two were engaged in a military campaign preventing the
Oghüz from traversing Gharjistan and threatening Ghur. Notably, Shah-i
Mashhad’s patron Jauhar Malik was not only the wife of Sultan Ghiyath
al-Din, she was also the sister of Saif al-Din, so that brother and sister were
Ghiyath al-Din’s paternal cousins.15
Given the altered routes of communication throughout central
Afghanistan in the modern day, both the Chisht and Gharjistan
complexes – not to mention the seemingly enigmatic but pivotal site of
Jam-Firuzkuh (cf. also Chapter 3) – have all received much less scholarly
attention than they deserve. Moreover, the majority of recent and ongoing
studies – including this one – have had to rely largely on archaeological
and photographic documentation conducted decades earlier. Nevertheless,
re-examination of the structures’ architectural fabric, decorative iconog-
raphy, and inscriptional programs provides invaluable insight into the
coalescence of the Firuzkuh branch of the Shansabani confederacy, and
the alacrity of this newly coalescing elite – despite their origins in the
nomad–urban continuum – in joining the fray of architectural patronage
to achieve political ends.
Until recently, both the Chisht and Gharjistan complexes appeared to
have existed in splendid isolation, as monumental and heavily inscribed
behemoths suddenly rising above expansive swaths of cultivated fields or
mountainous landscapes, which were crossed here and there by through-
ways plied principally by nomads (Figure 4.4).16 Such apparent isola-
tion would undermine the substantial investment of labor and resources
required in building what were originally large and rambling complexes;
they appear to have been intended for habitation, just as their painstak-
ingly planned and executed inscriptions were intended for reading, at least
by the learned among their residents and passers-by.
In the early 2000s, the mapping and documentation of structural
remains from Herat toward Chisht, extending eastward and flanking the
Hari Rud, turned up residual traces of a very different historical landscape.
To be sure, many remnants of fortified towers continue to be difficult
to date, presenting problems not dissimilar to those in dating the struc-
tural remains in southern Ghur and west of Bamiyan (cf. Chapters 2
and 3). However, the formal characteristics and modes of construction of
several other ruins along the Herat–Chisht routes indicate that these were
either built – or perhaps already in use – during the period of Ghaznavid,
Saljuq (late tenth–mid-twelfth centuries) and later political presences in

Figure 4.1 Remains of Shah-i Mashhad madrasa, c. 1165–1175 CE, Gharjistan, Badghis Province, Afghanistan. © Photograph Bernt Glatzer c. 1970.
14/09/21 5:39 PM
the area.17 These newly mapped architectural remains essentially outline

a network of large and small structures marking passages along the Hari
Rud, seemingly punctuated by nodes of culmination, quite possibly like
the originally sizeable complex at Chisht (discussed further below). While
similar exploration, mapping, and documentation are still to be conducted
for Shah-i Mashhad and its vicinity of the lower Murghab River (Figures
4.2 and 4.3), re-conceptualizing both of these magnificent monuments as
integral nodes in networks facilitating travel and communication appears
to be historically more accurate; moreover, such a contextualization helps
to explain the complexes’ lavishness and – above all – their prominent
epigraphic programs.
Shah-i Mashhad (Gharjistan)

In his seminal work of 2009, F. B. Flood agreed with and adduced further
evidence toward the scholarly consensus that Shansabani adherence to and
patronage of the Karramiyya continued unfaltering through ah 595/1199
ce, “when the sultans of Ghur abruptly terminated their association with
[them], embracing the orthodox madhhabs … of Islam instead.”18 He
cited four additional points in support, including the possibility that the
tomb–madrasa complex of Shah-i Mashhad – dated to ah 561–570/1165–
1175 ce, as noted above – was a Karrami institution (Figure 4.1). Another
point was the erection of the minar of Jam-Firuzkuh in c. ah 569/1174
ce (Figure 4.18), whose epigraphic program of all of Qur’an XIX (Surat
Maryam) has been convincingly interpreted as a Karrami proclamation
(cf. Appendix, Afghanistan VII: 5).19 Flood’s third evidentiary datum
was a luxurious, four-volume illuminated Qur’an, dated by colophon
to ah 584–585/1189 ce: the extensive commentary (tafsir) copied at the
end of each Sura was originally composed by the leading Karrami of
eleventh-century Nishapur, Abu Bakr ‘Atiq ibn Muhammad al-Surabadi
(d. c. ah 494/1101 ce). According to Flood, the commentary indicated that
this Qur’an “was among the manuscripts commissioned by the Ghurids
for a Karrami madrasa,” perhaps even Shah-i Mashhad. Finally, Juzjani
described Ghiyath al-Din’s “conversion” to the Shafi‘i madhhab as being
effected through a dream or vision that came upon the sultan sometime in
ah 595/1199 ce.20
However, there is equally compelling evidence, both material as well as
circumstantial, that the Firuzkuh Shansabanis patronized sects beyond the
Karramiyya long before ah 595/1199 ce, much earlier than Flood and others
have proposed. The Ghaznavids and Saljuqs – the Shansabanis’ predeces-
sors and contemporaries (respectively) in the region – had benefited from
fomenting competition among the various schools of legal–theological
thought while attempting to establish and maintain control over various
parts of Khurasan. The juridico-intellectual rivalry was the by-product of

198 Iran to India
cultural patronage, one of the spheres of activity identified as integral to

the actuation of Perso-Islamic kingship during the twelfth century (cf.
Chapter 1). Stoking inter-madhhab competition most often resulted in the
amassing of greater and more centralized power precisely in the hands of
rulers and their agents.21 In light of these established precedents of state-
craft, it seems only natural that the Shansabanis – an emerging elite with
a transregional presence in Gharjistan as early as the 1150s, followed by an
even more consequential entrée into Ghazna by the 1170s – would also
extend their patronage to other madhhabs precisely at the time of creating
and consolidating their bases of power. Ghiyath al-Din’s official shift from
Karrami to Shafi‘i affiliation in ah 595/1199 ce did not necessarily preclude
patronage of other madhhabs beforehand, particularly by members of the
sultan’s family.
While Shafi‘i and Hanafi theologians, jurists, and their followers
tended to dominate politico-intellectual culture in the urban centers of
eastern Iran, their influence did not extend with equal impact into more
remote areas.22 Since the time of Ibn Karram (d. ah 255/869 ce), the
Karramiyya were much more adept than their urban colleagues at travers-
ing the distance between intellectual–cultural centers and their hinter-
lands. According to Bosworth, Ibn Karram had proselytized in “Ghur,
Gharchistan, and the countryside of Khurasan, denouncing both Sunnis
and Shi‘is alike, appealing especially to the peasants of those regions.”23
He eventually entered Nishapur with Gharjistani adherents already in
tow.24 In the early eleventh century, Sultan Mahmud Ghaznavi evidently
first urged the Karramiyya (back) toward Ghur as he attempted to rid
Nishapur of at least one of the contentious madhhabs that had made
control of the city a constant balancing act among its prominent intelli-
gentsia.25 The Karramiyya’s famed austerity underpinned their successful
penetration of a difficult terrain like that of Ghur, relieved only by vil-
lages and mountainous strongholds rather than large commercial centers.
Although many among the Sunni madhhabs considered the Karramiyya to
be anthropomorphists – and thus outright heretical – the sect continued
to hold some sway in the less trafficked reaches of Khurasan and elsewhere
well into the late twelfth century.26
Among the agriculturalists, nomadic pastoralists, and other non-urban
populations in regions such as Ghur, the Karramiyya simplified Islamic
orthodoxy and preached direct and unschooled access to Islam. For
example, the Karramiyya were known among Khurasan’s more mainstream
jurists for considering verbal utterance of Shahada as tantamount to full
conversion. While the city-centered madhhabs vigorously criticized them
for these practices, the latter actually garnered the Karramiyya enduring
and loyal followers. Notwithstanding the orthodox madhhabs’ objections
to such simplifications,27 such an easy distillation of Islam was conceivably
more efficacious than rigidly complex teachings for the conversion of

relatively isolated groups far from well-traveled routes and in the hinter-
lands of Khurasani centers.28
Given the frugality, if not asceticism, for which the Karramiyya were
known, it seems unlikely that a monumental and lavishly decorated
madrasa complex such as Shah-i Mashhad would conform to their ideo-
logical and practical standards.29 Karrami proselytism was indeed carried
out through group activity, and was closely associated with residence in
and operation from communal lodgings or khanaqahs, which also served
as sites of instruction and other charitable work (e.g., feeding and shel-
tering the poor). But historical authors often used khanaqah, ribat, and
madrasa interchangeably: by the twelfth century, these structures were all
multi-functional, communal spaces intended principally for instruction,
at times including a tomb and/or a mosque or other place for prayer.30
Moreover, the term madrasa did not necessarily imply large-scale con-
struction: the increasing monumentality of madrasas throughout the
mid- to late eleventh century has in fact been seen as a development at
least partially attributable to the Saljuq push to control Khurasan through
patronage of select madhhabs.31 Thus, while textual sources referred to
Karrami institutions as madrasas in cities such as Nishapur, it is improb-
able that these were the towering edifices associated with the orthodox
madhhabs favored both by the ruling elites and the economically prom-
inent mercantile groups.32 Ultimately, an exquisitely illuminated Qur’an
could command ascetic spiritual fervor qua ritual object; its architectural
equivalent, however, would surely be perceived as alien to the austerity
that characterized Karrami preceptors and their principles, an austerity
that continued as one of the few constants over time, distinguishing their
modus vivendi from that of the other madhhabs.33
But eliminating a Karrami affiliation for Shah-i Mashhad brings us
little closer to definitively ascertaining the madhhab – or possibly more
than one34 – housed at the madrasa. The complex was designed and built
on a monumental scale and, even after much deterioration, retains traces
of an exquisite decorative program encompassing admirably executed
monumental inscriptions (discussed further below). These characteristics
emphatically point to the complex’s affiliation with the established legal–
theological schools favored in Khurasan, quite possibly the Hanafi and/
or the Shafi‘i madhhabs – the latter proposed by S. Blair.35 Nevertheless,
the orientation of Shah-i Mashhad, along with its extensive and unusual
epigraphic program, both lead to more questions than answers, and point
to multiple possibilities for the complex’s sectarian affiliation.
Shah-i Mashhad has a nearly north–south orientation (Figures 4.2 and
4.3), so that the whole complex deviates counterclockwise from the cardi-
nal directions of true south and west (180° and 270°, respectively) by only
a few degrees, possibly due to accommodation of the terrain – though
this is unverifiable without on-site study. Since 243.69° would be Shah-i

200 Iran to India
Figure 4.2 Shah-i Mashhad madrasa and Murghab River. © Google Earth 2018.
Figure 4.3 Shah-i Mashhad madrasa on Google Earth, detail with superimposed
plan (from Najimi 2015: 153). © David Thomas 2018.
Mashhad’s precise qibla – achieved by hypothetically rotating the complex

several degrees further counterclockwise – M. Casimir and B. Glatzer
declared this “distinct deviation … a problem which … has not been suffi-
ciently examined.”36 Thankfully, during the intervening years since Casimir
and Glatzer’s initial rediscovery of and publications on the site, scholars
such as David King have examined this “problem” extensively, providing

invaluable insight into why not only Shah-i Mashhad but many contempo-
raneous mosques and other religious structures, ranging from the Maghrib
through Central Asia, were not always precisely oriented toward Mecca.
It is well known that, by the second millennium ce, the mathe-
matical specialties of trigonometry, astronomy, and stellar calculation
were sophisticated fields of study throughout the Islamic world.37 But,
pre-modern qibla determination had never been a purely “scientific”
matter. Architectural qibla orientations in particular frequently resulted
from “folk astronomy”: this was, in brief, a varying combination of the
regional or local prevalence of a given madhhab; any local awareness of the
Ka‘aba’s own complex cosmography, as well as its orientations according
to pre-Islamic wind theory; and topography and/or pre-existing urban
development. The preferred qiblas of all the madhhabs and associated
or independent smaller sects are not known – much less the fascinating,
socio-historical reasoning behind them.38 For present purposes, however,
it is fortunate that the preferred qiblas of the Hanafis and the Shafi‘is, the
two most prevalent madhhabs in Khurasan, are known: while the Shafi‘is
subscribed to a qibla due south (180°), the Hanafis subscribed to a qibla
due west (270°).39 Nevertheless, in the absence of an exhaustive catalog of
architectural orientations across the Islamic world through the thirteenth
century, it is still unclear how closely or consistently these prescribed qibla
orientations were actually followed,40 even though they were favored by
two of the most prominent Sunni madhhabs.
The orientation of Shah-i Mashhad along a near exact north–south axis
could accommodate either or both of the prevalent Khurasani madhhabs’
preferred qiblas. But the complex’s already deteriorated – and worsening –
state of preservation (Figures 4.1 and 4.4) impedes more detailed obser-
vations. Based on an extant but previously unread naskh inscriptional
fragment on the monumental southern entrance, A. W. Najimi proposed
that the surviving domed chamber at the quadrangular compound’s
southeastern corner (Figure 4.5) contained the tomb of a deceased
male personage, quite possibly the slain Shansabani sultan Saif al-Din
(d. ah 558/1163 ce). Although the surviving fragment does not name Saif
al-Din specifically, the association is tenable based on the sultan’s death
having occurred in the general vicinity of the madrasa, shortly before its
foundation.41 Furthermore, we recall that Saif al-Din was the brother of
Shah-i Mashhad’s patron, Queen Jauhar Malik (see above). The poor
state of preservation, however, makes it impossible to determine whether
there was a mihrab in the chamber. And the outright disappearance of
the rest of the large compound precludes knowing if one or more prayer
areas were housed elsewhere within it, much less their qibla orientations.
Nevertheless, it is noteworthy that Shah-i Mashhad’s overall north–south
orientation accommodates the qiblas prescribed by both the orthodox
madhhabs dominant in Khurasan.

202 Iran to India
Figure 4.4 Shah-i Mashhad madrasa, south façade. © Photograph Bernt

Glatzer c. 1990?
Figure 4.5 Shah-i Mashhad madrasa, possible tomb of southeast corner.

© Photograph Bernt Glatzer c. 1970.
Ever since the rediscovery of Shah-i Mashhad in 1970, its epigraphy and
extensive ornament have commanded as much attention as its structural fea-
tures. The inscriptions’ calligraphic styles range from knotted and interlock-
ing Kufic, to elongated naskh and thuluth against lush floral backgrounds

(see Appendix, Afghanistan

IV). Indeed, scholars have
come to consider the com-
plex’s ornamental–structural
wholeness to be an over-
arching characteristic of
Shansabani foundations, so
that “[i]n Ghurid buildings,
architecture was the struc-
ture and decoration; and
decoration and structure
were the architecture.” The
content, calligraphic styles,
and variety in materials and
techniques of the madrasa’s
epigraphic program have
long been the subjects of
discussion and debate, with
there being scholarly con-
sensus only on the difficulty
of discerning the program’s
presumably multiple and
complex meanings.42
Casimir and Glatzer
deciphered the first four
verses of Surat al-Fath (Q.
XLVIII) in Kufic script on
the surviving eastern end of
Figure 4.6 Shah-i Mashhad madrasa: south façade, detail of
Shah-i Mashhad’s monu-
inscription. © Photograph Bernt Glatzer c. 1970. Courtesy of
mental south façade, whose IsMEO, Rome.
imposing central portal
carried the Arabic foundation inscription in an even more ornate, plaited
Kufic (Figures 4.1 and 4.6).43 The original symmetry of the façade makes it
plausible that the western flank continued with at least three more of the
Sura’s following verses (Q. XLVIII: 5–7).44 Much more recently, Najimi
deciphered or reinterpreted five previously unread or differently inter-
preted inscriptions on Shah-i Mashhad’s southern and northern façades,
in addition to the cursive inscription referring to a deceased male’s tomb
(cf. above).45 The other five include the takbir of tashriq46 in thuluth script
on the inner eastern face of the main southern portal (Figure 4.7), below an
Arabic Kufic inscription naming, presumably, the master builder, or the one
who “completed this building, the servant Ahmad ibn Mahmud” (Figure
4.8). On the extant pier of the northern portal of the complex, Najimi deci-
phered verses 22–24 of Surat al-Hashr (Q. LIX) in Kufic script, as well as a

204 Iran to India
Figure 4.7 Shah-i Mashhad madrasa, southern monumental entrance, interior

with inscription. © Photograph Bernt Glatzer c. 1970.
naskh inscription of a Hadith apparently of Shi‘i association (Figure 4.9).47

The presence of verses from Surat al-Fath is unusual in the eastern Islamic
lands,48 and Najimi’s new readings and reinterpretations make all the more
poignant the lingering questions about the madrasa’s sectarian affiliation(s),
the patron’s role in the complex’s conception, as well as the intended audi-
ence(s) of the inscriptions. The import of juxtaposing all of these inscrip-
tions in an admittedly complicated epigraphic program is discussed below.
But first, it should be noted that many factors rendered difficult both
reading and also fully comprehending epigraphic programs, not only at
Shah-i Mashhad but also at the other architectural sites attributed to
this first phase of elite patronage within the territories of the Firuzkuh
Shansabanis (cf. Volume II for late twelfth-century architectural patron-
age). Even though Shansabani-patronized inscriptional programs were
almost exclusively in Arabic,49 their varying and ornate scripts often hin-
dered rather than facilitated reading. Furthermore, the inscriptions varied
greatly in content, ranging from Qur’anic verses, historical and funerary
inscriptions, Hadith, and possibly even didactic texts – the last exemplified
in Shah-i Mashhad’s takbir of tashriq. All of these challenges were prob-
ably as pronounced or perhaps even more so when the buildings’ programs
were complete. Thus, all considerations of the overall design, composi-
tion, and readership of the inscriptions on monuments throughout the
Firuzkuh Shansabanis’ realms must necessarily proceed from the premise
that readers of the inscriptions were extremely few.50

Figure 4.8 Shah-i Mashhad madrasa, south façade, detail of inscription.

© Photograph Bernt Glatzer c. 1970.
In fact, it is not unreasonable to propose that the Shansabani elites

themselves were unable to read inscriptions in their entirety, much less
determine the minutiae of their content or the overall compositions of
the epigraphic programs. Two broader historical realities bolster the idea
that literary documentation and literacy in general – as understood in the
modern sense – were not common throughout Afghanistan during our
period of focus. First, the historical trajectory of the Shansabani elites and
their emergence from – and likely some maintenance of – pre-imperial tran-
shumance and/or nomadism made for a selective knowledge of the “Great
Traditions” of Islamic juridical and theological doctrines, and perhaps
even Islamic religious practices.51 Among the Shansabanis’ forebears –
spanning an array of lifestyles across the nomad–urban continuum (cf.
Introduction and Chapter 2) – literacy was not always a requisite, nor was
it necessarily widespread: the past and present of many self-identifying
groups, for example, were frequently recorded and preserved in collective
memory and transmitted orally across generations, captured in written
form only a considerable time afterward, if at all.52
We have seen that the vast differences between urban center and hinter-
land were instrumental in the version of “Islam” that was propagated: while
some Karrami legal and theological scholars were known to have been the
erudite interlocutors of other madhhabs in Khurasan’s great emporia,53
their comrades’ proselytism of a thoroughly simplified version of Islam
among non-urban inhabitants – embodied in the Karrami tenet that mere

206 Iran to India
recitation of Shahada was

tantamount to full Islamic
conversion (see above) –
likely bypassed even the
limited, mnemonic liter-
acy of other pre-modern
Islamic populations.54
Such a loophole would
have been all the more
necessary among people
of rural, nomadic, and/
or transhumant lifestyles,
largely living beyond the
urban settlements of what
is now Afghanistan. Thus,
royal Shansabani involve-
ment in the sophisticated
epigraphic and decorative
programs of their monu-
mental commissions was
conceivably secondary
to that of the literate,
even well-read advisors
and eventual users of the
buildings. Execution of
the designs by generation-
ally trained artisans, spe-
cialized in various aspects
of building and decora-
tion, surely contributed
a formative element as
Figure 4.9 Shah-i Mashhad madrasa: north entrance, detail well.55
of inscription. © Photograph Bernt Glatzer c. 1970.
This is not to say that,
as patrons of “civic” architecture, the Shansabani sultans and their families
and associates made absolutely no mark on their foundations or the lat-
ter’s complex epigraphic programs. But it must be borne in mind that at
least the Shansabani family members – if not also their nobles and other
associates – certainly differed from their most proximate contemporaries.
The Ghaznavids, for example, whose scions had been thoroughly educated
in Sunni Islam and steeped in Persianate culture, were surely discerning
patrons probably even intervening in architectural and epigraphic details.56
The Shansabanis, though also preoccupied with consolidating and expand-
ing an imperial presence (like the Ghaznavids), nonetheless had come to an
overt embrace of Islamic socio-religious mores and Persianate ceremonial

forms in relatively recent times; and it was argued in Chapter 2, that they
did so in large part for the purposes of projecting an imperial image. In
light of their historical circumstances, the Shansabani elites were surely
little interested in the subtleties of epigraphic messaging. Beyond aspiring
to monumental precedents – as proposed below – the details of Shah-i
Mashhad’s complicated inscriptional program were surely left to the more
learned, future inhabitants and frequenters of their monumental founda-
tions, viz. theologians, jurists, their students, visiting intelligentsia, and
literate passers-by.
In the specific case of Shah-i Mashhad, the great variety in content
of the madrasa’s surviving inscriptions points to expert and intentional
epigraphic choices, probably made by its well-read dwellers rather than
its royal patron(s). Although no definitive conclusions are forthcoming
due to the disappearance of the majority of the complex and its inscrip-
tional program, what remains seems to encompass contradictory trends:
while a few of the inscriptions follow epigraphic developments in the
eastern Islamic lands and beyond, others notably diverge from them. The
northern entrance’s verses from Surat al-Hashr (Q. LIX) (Figure 4.7),
for example, were not uncommon, particularly in funerary contexts in
the Levant, Egypt and the Maghreb from the first two centuries of the
Hijira, and in greater Iran as of the eleventh century.57 The presence of
these verses further supports the probability of a tomb within the madrasa,
perhaps even that of the Shansabani sultan Saif al-Din (discussed above).
By contrast, the presence of a possibly Shi‘i Hadith on the northern
entrance (Figure 4.9), and the takbir of tashriq on the southern entrance
(Figure 4.7), are both unusual. Quotations from Hadith were rare in the
eastern Islamic lands, there being but one documented instance from the
early eleventh century,58 and no known quotations of the takbir at all. Shi‘i
groups had not often suffered persecution since Ghaznavid ascendancy in
greater Khurasan, particularly if their activities did not interfere with the
political elites.59 But, even though a Hadith from the Shi‘i canon on the
madrasa would not necessarily be remarkable, it seems improbable that
Shi‘i instruction would take place there given the Shansabanis’ Sunni
affiliation. It should also be noted that, in general purport, the quoted
Hadith is not dissimilar to Hadith 425 in Sahih of al-Bukhari (ah 194/810
ce–ah 256/870 ce), thus generally accepted by the Sunni madhhabs.60 The
takbir on the southern entrance’s interior is currently inexplicable, except
possibly as a didactic inscription. Ultimately, the madrasa’s surviving
epigraphic program not only intimates a sectarian ecumenicalism, it may
also point to localized developments that will become clearer after analysis
by specialists in Islamic theology and/or further archaeological work.
While much of Shah-i Mashhad’s inscriptional program doubtless
emerged from the input of the ‘ulama’ and other learned individuals who
would be its future residents, one prominent part of the epigraphy could

208 Iran to India
Figure 4.10 Ghazni, minaret of Mas‘ud III (1099–1115), with citadel in

background. © Photograph Josephine Powell c. 1960. Harvard Fine Arts Library,
have been the expression of the royal patrons’ preference: the initial verses
of Surat al-Fath (Q. XLVIII) as the principal inscriptions on the madrasa’s
main façade along the complex’s southern perimeter (Figure 4.1). Verses
from this Sura were extremely rare in monumental epigraphy anywhere in
the Islamic world, with their first documented occurrence in the eastern
Islamic lands on the minar of Mas‘ud III (r. ah 492–509/1099–1115 ce)
at Ghazna (Figures 4.10 and 4.11).61 This memorable and towering archi-
tectural work had been among the structures that escaped Sultan ‘Ala’
al-Din Husain’s sacking of the city after his defeat of Bahram Shah in
ah 544–45/1150 ce – the occasion that earned him the sobriquet Jahan-suz
(cf. esp. Chapter 1) – only about fifteen years before the foundation of
Shah-i Mashhad.62
Pinder-Wilson convincingly proposed that the Ghazna minar’s inscrip-
tional program had been inspired by the resumption of successful cam-
paigns to northern India during the reign of Mas‘ud III in the early twelfth
century, after a lapse of about a half-century during which the Ghaznavid

Figure 4.11 Ghazni, minarets of Ghazni: minaret of Mas‘ud III (1099–1115),

middle section, detail. © Photograph Josephine Powell c. 1960. Harvard Fine Arts
house had seen turbulent internal strife.63 Mas‘ud III’s re-initiation of the
profitable and propagandistically important Indian raids reclaimed some
of the Ghaznavids’ lost imperial glory, and also provided the sultan with
the occasion to erect a towering structure commemorating what were
essentially his victories.64 The inscription of Surat al-Fath – unprecedented
in its entirety of twenty-nine verses65 – was in fluid cursive against a floral
background, contained within a band meandering around quadrangular
panels occupying the minar’s remaining surface areas (Figure 4.11). But
equally notable was the royal appropriation of a visual device traditionally
reserved for religious epigraphic content. Interlocking (ma‘aqali) Kufic,
ingeniously forming a grid of words, was not uncommon on funerary stelae
and architectural surfaces (eleventh–twelfth centuries); but the content of
these inscriptions had virtually always been religious in nature, such as the
names of the first four khalifas, or doxologies. However, the horizontal
band of square panels near the base of the minar – its lower positioning
making it more legible – was filled with the sultan’s kunya and some of his

210 Iran to India
regnal titles.66 Thus, the monumental minar was a victory tower in form
as well as decoration.
Although the full circumstances of Shah-i Mashhad’s construction are
not known, it is safe to say that this complex was much more than – or
at least quite different from – a victory tower. Surat al-Fath was likely
not inscribed in its entirety here nor were the quoted verses in cursive,
as at Ghazna, where they formed a contrast to the less legible, ornate, or
interlocking Kufic of the regnal titles. At Shah-i Mashhad, however, the
selected Qur’anic verses from Surat al-Fath and Surat al-Hashr survive in
the generally less legible Kufic.67 In fact, the verses from Surat al-Fath on
the south façade actually appear subordinate to the much more ornate and
larger, plaited Kufic of the foundation inscription of the central arch, as if
building up to the façade’s epigraphic apex (Figures 4.1 and 4.6).
Given the Firuzkuh Shansabanis’ imperial aspirations – and, argua-
bly, their furthering of these aspirations by patronizing Khurasan’s more
prominent Sunni madhhabs – the inclusion at Shah-i Mashhad of verses
from Surat al-Fath appears to be an indexical citation of their reputed
predecessors the Ghaznavids. Rather than an exclusive “desire to convert
the heathen” or to commemorate the recent military victory against the
Oghüz,68 it was conceivably the emulation of an iconic monument and its
dynastic context that also led the Shansabani queen Jauhar Malik, Shah-i
Mashhad’s patron, and/or her spouse Sultan Ghiyath al-Din to choose
verses from Surat al-Fath as part of the madrasa’s epigraphic program.
Ultimately, Shah-i Mashhad accomplished multiple aims for the Firuzkuh
Shansabanis: the patronage provided them with an architectural “debut”
as patrons of madhhabs that were prominent across Khurasan. At the same
time, such an epigraphic proclamation also invoked the imperial echelon
of the Ghaznavids, who were once illustrious but now were on the wane,
superseded by the Shansabanis.
Chisht
Given the proximity of the two surviving domed structures at Chisht
(Figures 4.12), it is probable that they originally formed part of a larger
complex.69 This foundation adduces even more evidence in favor of the
activity of madhhabs other than the Karramiyya in Shansabani-ruled lands
prior to the end of the twelfth century.
On the interior surface of the western structure, the cursive inscrip-
tional frieze at the base of the dome (Figure 4.12 [left structure]; Figures
4.13 and 4.14) records in Arabic the renovation (tajdid) of the building “in
the days of Shams al-Dunya wa al-Din” (see also above). There is little
doubt that this personage was the Shansabani sultan Ghiyath al-Din, since
the surviving remainder of the inscription continues with his kunya Abu
al-Fath, his ism Muhammad, and nasab of ibn Sam.70 The accompanying

Figure 4.12 Chisht, mausoleums (or madrasa/mosque complex), ruins: Mausoleum

A, exterior, southern façade. © Photograph Josephine Powell c. 1960. Harvard Fine
Kufic inscription (Figure 4.15) furnishes the date of ah 562/1167 ce in

Persian.71 Chisht’s less preserved, northeastern structure (Figure 4.12 [right
structure]) also has an Arabic inscription with Ghiyath al-Din’s correct
titles, but in foliated Kufic and, notably, as they eventually appeared at the
imposing jami of Herat at the end of the twelfth century.72
By considering the southwestern domed structure’s naskh and Kufic
inscriptions together, the seemingly egregious anachronism of the sultan’s
pre-regnal laqab of Shams al-Dunya wa al-Din – applicable only prior to ah
558/1163 ce – in the cursive text may present us with a chronological clue,
not only for the structure but also the overall complex. It is worth consider-
ing that the southwestern structure was completed before the northeastern
one, and prior to ah 558/1163 ce, or at least before wide circulation of the
sultan’s regnal laqab of Ghiyath al-Din.73 The contents of the southwestern
structure’s naskh inscription, then, would also have been composed at this
earlier date, thus still using the sultan’s earlier name. The separate Kufic
inscription in the same structure, however, lists the date of ah 10 Jumada
I 562/4 March 1167 ce, but only after a long series of Qur’anic verses
rather than the description of an event. Given that the cursive and Kufic
inscriptions are separate, it is possible that the latter does not refer to the
above-mentioned renovation (tajdid), but rather to some other, as yet not
fully known, intervention in the structure or complex as a whole.

212 Iran to India

A, interior, corner pillars and two adjoining walls with inset arches, with bands
of inscriptions following outline of upper arches and running horizontally across
midpoint of arches. © Photograph Josephine Powell c. 1960. Harvard Fine Arts
Like Shah-i Mashhad, the Chisht structures are also oriented along
a north–south axis with a slight deviation, but here clockwise, with the
buildings’ western walls again facing just perceptibly off true west (Figure
4.16).74 While the nature of the original complex is difficult to discern
with certainty, parts of the epigraphic program and other features of both
domed structures provide clues as to what was likely a site with multiple
functions. Tellingly, the more deteriorated northeastern building (Figure
4.12 [right structure], Figure 4.16) still preserves a mihrab on its western
wall (Figure 4.17); in its original state the mihrab was probably plastered
and covered with lavish molded stucco decoration, remnants of which
are still in situ above the arch. The large pointed-arch motifs diagonally
above and on either side of the mihrab not only echo the latter’s form,
but the curves of the arches contain bands framing the above-discussed
foliated Kufic inscription with Ghiyath al-Din’s titles. Smaller medallions


A, interior, inscription, detail. Powell: “Detail of [inscription] which has been read
to me, ‘In the days of the kingdom of Al Mozafar Almansour …’”*
*Cf. Appendix II, Afghanistan III:4, below, for an updated reading of
this inscription.
within the large corner arches contain the names of the first four khalifas,
the Prophets, Shahada, and doxologies. The smaller, paired-arch motifs in
between contain doxologies.75
Although only a small portion of the Chisht complex’s epigraphic
program survives – not dissimilar to Shah-i Mashhad – it is noteworthy
that the available inscriptions as well as the complex’s formal characteris-
tics point toward a funerary function for the site, probably serving other
purposes as well. The domed chamber layout of the northeastern structure,
discussed above, strongly suggests that it was a tomb with a mihrab – a
not uncommon architectural typology in the eastern Islamic lands by the
later twelfth century.76 Additionally, the western domed structure, while
apparently not having a mihrab, provides ample epigraphic indications of
a funerary association through its interior inscriptions. The cursive inscrip-
tion beginning in the southeastern corner with the anachronistic mention

214 Iran to India
of Shams al-Dunya wa al-Din (Figures 4.13 and 4.14), continues with parts
of verses 23–24 of Surat al-Hashr (Q. LIX), which was also part of the
epigraphy on Shah-i Mashhad’s northern entrance, and whose frequent
presence on epitaphs, and funerary stelae and architecture has already been
noted (cf. notes above).
The western structure’s Kufic inscription (Figure 4.15) contains the
common verses 255–257 from Surat al-Baqara (Q. II), among which is the
famous Throne verse (Q. II: 256). The text continues with verses 18–19 from
Surat Al ‘Imran (Q. III), “the most popular Koranic text on monuments
from early Islamic Iran.”77 As if by deliberate contrast, the Kufic inscription
progresses with verses 1–3 from Surat al-Ikhlas (Q. CXII), which were
undocumented in Blair’s inscriptional corpus of the eastern Islamic lands
from the ninth through early twelfth centuries, but common particularly
in funerary contexts (mainly epitaphs) from the ninth century onward in
Egypt, and also in Iran and the Red Sea coast of northeast Africa.78
Furthermore, a consideration of the directional orientation of the
Chisht complex (Figure 4.16) and an examination of the site’s broader
historical context, together provide a fascinating if complicated picture of
multiple madhhabs active in the region, probably through indirect means.
First, the orientation of the mihrab within Chisht’s northeastern structure
could hint at a Hanafi affiliation for the complex, or at least a portion of
it. As discussed in relation to Shah-i Mashhad (cf. above), pre-modern
qibla orientations were calculated largely by means of “folk astronomy,”
in which the preferred qiblas of regionally dominant madhhabs were
among the determining factors. While the Shafi‘is were known to use a
south-oriented qibla, Hanafi qiblas tended to be oriented west. The Chisht
complex’s nearly north–south axis could have accommodated either or
both orientations, similar to the layout of Shah-i Mashhad. But at Chisht,
the surviving mihrab in the northeast structure has a westerly orientation,
thereby seeming to dispel any ambiguity in favor of a Hanafi affiliation.
Finally, it is possible that, particularly if there was a Hanafi-affiliated
presence at the Chisht complex, other madhhabs were also active in the
vicinity, if not directly then through a type of surrogate representation.
Chisht was certainly within the ambit of the Khurasani commercial center
of Herat, only 150 km to the west, which in turn was well connected with
Nishapur and other such great emporia beyond. But as noted earlier in
this chapter, the direct influence of the city-based madhhabs was consider-
ably diluted in the hinterlands, precisely where the more intrepid activists
among the Karramiyya had identified proselytizing opportunities.
The Karramiyya were not alone, however, in their unorthodox pros-
elytizing methods in both urban and hinterland milieus (cf. above in
this chapter). Charismatic spiritual masters or Sufis (also pir, murshid) –
identifiable by a characteristic woolen cloak (suf) or other mark of poverty –
were also gathering followers devoted to their teachings, in cities such as


A, interior, arch, with inscription, detail. © Photograph Josephine Powell c. 1960.
Nishapur and Herat, and in less traveled reaches.79 Although the follow-
ers of Sufi masters originated in various rungs of society – ranging from
lay devotees of prosperous mercantile backgrounds through artisans and
craftspeople, as well as the indigent poor – true renunciants (murid)
would eventually coalesce as orders (tariqa) around their pirs, coming to
be associated with specific khanaqahs. There was great variation among
Sufi tariqas in both geographical reach and longevity, with some attain-
ing hereditary followings (silsila) able to spread the pir’s teachings far and
wide through a process of associative fragmentation: new silsilas could
be established, particularly in areas at a distance from their spiritual
order’s origin, yet remain affiliated, even loosely, with the memory of
the initial pir and the spiritual teachings attributed to him.80 The epon-
ymous Chishtiyya was just such a Sufi order, having associations with
the locality of Chisht as early as the ninth century, but attaining its

216 Iran to India
Figure 4.16 Chisht on Google Earth, October 2018.

B, interior, mihrab, with decorative stucco work. © Photograph Josephine Powell

greatest followings and long historical presence throughout the Indian

subcontinent.81
Given the migration of the order’s pir, Mu‘in al-Din Sijzi Chishti (ah
536–633/1142–1236 ce), to northern India probably by the 1180s, when the
Ghazna Shansabanis took Lahore (see Chapter 5), it is unlikely that the
monumental remains of the Chisht complex housed the Sufi order; their
khanaqah would have been housed by another, probably more humble
structure.82 Moreover, the complex’s at least partial Hanafi affiliation – as
argued above – would further make a link with the Chishtiyya or other
Sufis precarious. Although spiritual teachings definable as mystical or
“Sufi” encompassed enormous variety, in general it can be said that they
were of a very different order than theological–juridical doctrines. Being
much more personal, performative, and “ascetic–mystical”83 as spiritual
paths, Sufi teachings could not uniformly be associated with any of the
orthodox madhhabs, and in fact could have gravitated toward Shi‘ism and
particularly its Isma‘ili expression.84
The decidedly independent or unorthodox expressions of Islam associ-
ated with Sufism overall made it all the more likely to appeal to non-urban
populations (not unlike Karrami teachings). Nevertheless, Bulliet noted
that, certainly in eastern Khurasan, “mysticism was almost non-existent
among the Hanafis … and most directly associated with the Shafi‘i party”
at least through the early twelfth century.85 While the evidentiary lacunae
make all observations conjectural, this brief digression on the Sufism
burgeoning in Afghanistan during the twelfth–thirteenth centuries sup-
ports the idea that at Chisht also, where the Shansabanis were epigraph-
ically acknowledged, madhhabs and spiritual movements other than the
Karramiyya could have benefited – probably via indirect avenues – from
the Shansabanis’ political presence and perhaps even patronage.
The Minar of Jam-Firuzkuh

The preceding observations regarding the multiplicity of madhhabs in areas
ostensibly within the Firuzkuh Shansabanis’ purview in no way refute the
Karrami affiliation of the minar at Jam-Firuzkuh (Figure 4.18). If anything,
the foregoing scenario lends additional support for the minar’s interpreta-
tion as a monumental vehicle for Karrami expression, at the very heart of the
Firuzkuh Shansabanis’ sard-sir “capital.” However, the minar’s epigraphic
program was arguably the Karramiya’s swan song, serving as an indica-
tion of their curtailed sphere of influence – increasingly focused on the
Shansabanis’ royal summer encampment and the local population – and
their last response to these straitened circumstances before the Firuzkuh
Shansabanis officially espoused the Shafi‘iyya in ah 595/1199 ce.
Sourdel-Thomine’s86 definitive re-dating of the Jam-Firuzkuh minar
from ah 590/1193–1194 ce to ah 570/1174–1175 ce certainly divests it of

218 Iran to India
Figure 4.18 Jam, minaret of Jam (c. 1180), distant view along valley
floor. © Photograph Josephine Powell c. 1960. Harvard Fine Arts Library,
any association with the Shansabanis’ campaigns into the north Indian
plains: two unsuccessful forays took place in ah 573/1178 ce and ah
587/1191–1192 ce, and it was not until ah 589/1192–1193 ce that the
Shansabanis gained a foothold in the north Indian duab (cf. also Chapter
1 and 6, and Volume II). Nevertheless, the minar’s architectural form
and typology undeniably connect it to the minars still standing in the
Persianate world, in the territories affiliated both with the Saljuqs and
the Ghaznavids – the latter particularly at the erstwhile Yamini capital.
The Shansabanis’ known emulation of their Ghaznavid predecessors –
discussed above – strongly suggests that the Jam-Firuzkuh minar also
served to commemorate an important military triumph. Furthermore,
the presence of all of verse 13 and the beginning of verse 14 from Surat
al-Saf (Q. LXI), promising imminent victory (Ar. fath) to those who
believe in Allah, also supports the proposal that commemorating a victory
was a significant – if not the principal – impetus behind the minar’s

construction.87 Although placed toward the pinnacle, the verses’ ren-

dition in square Kufic allowed for some legibility, at least of certain
words, permitting even the less literate among its beholders to recognize
the verses.
Indeed, the ideal occasion for commissioning such a monument had
presented itself only the year before: in ah 568/1173 ce, a joint push by
both of Baha’ al-Din’s sons finally routed the Oghüz encamped at Ghazna,
resulting in the Shansabanis’ definitive occupation of the city, the crown-
ing of Shihab al-Din as Sultan Mu‘izz al-Din, and the establishment of
a third Shansabani lineage there – discussed in detail in Chapter 5. Such
a significant victory would clearly warrant a monumental commemora-
tion like the splendid minar.88 Moreover, the location of the minar at
Jam-Firuzkuh, rather than at Ghazna, distinguished it from its architec-
tural predecessors and signified a Shansabani appropriation of the form
and its signification. Ultimately, the minar plausibly played the role of
a victory “trophy”: although Shihab al-Din was crowned Sultan Mu‘izz
al-Din at Ghazna, the minar’s location at Jam-Firuzkuh – and its concom-
itant absence at Ghazna – would serve as a constant (if dyadic) reminder of
the familial hierarchy wherein the Firuzkuh Shansabanis sought to be the
senior-most lineage of a territorially expanding confederacy.
The analysis of Shah-i Mashhad’s extant inscriptions earlier in this
chapter revealed not only the significance of the epigraphic programs on
Shansabani-patronized architecture, it also underscored these patrons’ cir-
cumscribed role in determining their content. The distinctive (but com-
plementary) purviews of the Shansabanis as architectural patrons, vis-à vis
the theological, legal, and literary intelligentsia they also supported, does
much to explain the most prominent component of the Jam-Firuzkuh
minar’s epigraphic program: all of Surat Maryam’s (Q. XIX) ninety-nine
verses, meandering vertically and delimiting decorative panels all around
the minar’s lowermost cylindrical shaft (see Appendix, Afghanistan
VII: 5).
J. Sourdel-Thomine and F. B. Flood convincingly argued that the
Karramiyya still installed at Firuzkuh likely chose this Sura as the minar’s
epigraphic centerpiece, aiming it at their theological rivals rather than
their Shansabani patrons. The Shansabani elites’ increasing connections
with the more prominent madhhabs of Khurasan – as argued throughout
this chapter – would have warranted just such a gesture. The Shansabani
conquest of Hanafi Ghazna, and the towering commemoration of the
victory at Firuzkuh, served as the Karramiyya’s monumental refutation
of the charges of anthropomorphism that the more prominent Sunni
madhhabs had leveled against them, beginning as early as the eleventh
century.89 The Shansabanis’ lack of engagement with the finer points of
theology would have made one Sura as acceptable as any other. But for the
Karramiyya, such a riposte to their detractors was a matter of political and

220 Iran to India
economic survival, as they attempted to hold on to the dwindling support

of their sole remaining royal patrons.
The preceding treatment of early buildings in the ambit of the Firuzkuh
Shansabanis has aimed to highlight the manifold significances of archi-
tectural patronage in building empire, as much for elites emerging from
nomadism and/or seasonal transhumance as for all other power-brokers.
Indeed, the Shansabani elites’ limited conversance with literary cultures
beyond recognizable Qur’anic phrases or verses, and other common
Arabic doxologies, also resulted in an epigraphic corpus devoid of the
types of poetic, political, and other inscriptions in Persian that graced
(for example) Ghaznavid complexes.90 Nevertheless, by commissioning
a monumental public structure such as the Shah-i Mashhad complex,
leaving at least an acknowledged presence at Chisht, and erecting the
Jam-Firuzkuh minar, the Firuzkuh Shansabanis were simultaneously
extending patronage to the various madhhabs and spiritual movements
vying for greater politico-economic prominence throughout Khurasan.
Perhaps most importantly, in the process of patronizing “civic” or other-
wise landmark architectural sites – emphatically not palaces – the ambi-
tious Firuzkuh Shansabanis were literally and figuratively building their
own bases of power. During the second half of the twelfth century, their
projects daringly began to encroach on the well-plied networks of trade
and communication radiating from the cosmopolitan emporia of Herat
and Nishapur (cf. esp. Volume II).
Patrons and Patronage
The foregoing analysis of the more firmly attributable built traces of the
Firuzkuh Shansabanis began with some seminal questions regarding
architectural patronage. We interrogated what monumental building ulti-
mately meant, what it accomplished especially for elites emerging from
transhumant/nomadic lifestyles and economies and going on to create
an empire. It is equally important to examine the overall relationship
between patron and outcome, that is, historically, the degree to which the
preferences of, and input by, the one(s) “paying the bills” impacted the
architectural result. There is surely no unitary answer to the last question;
investigators across disciplines would be hard-pressed to craft a convinc-
ing, universally applicable characterization of the relationship between all
patrons and their projects, even within a specific region and timespan.
These considerations are all the more relevant for the Shansabanis,
for at least two reasons. First, we have seen throughout Chapter 3, and
this one that the plural “Shansabanis” actually encompassed essentially
independent lineages, each responding to and shaping disparate cultural
geographies that ranged from Bamiyan through Ghur, and eventually

Zabulistan–Zamindawar (cf. Chapter 5). Second, given the Shansabanis’

origins in the nomad–urban continuum of seasonally transhumant and/
or nomadic lifeways, the fissure between patron and patronage becomes a
particularly informative space to explore.
Blair rightly noted that, “[i]n both form and intent, Ghurid architec-
ture continues that of the Ghaznavids.”91 However, much needs to be
differentiated in the processes of how these two architectural corpora came
to be. We have established elsewhere in this volume that the Shansabanis
can and should be distinguished from their immediate predecessors the
Yamini–Ghaznavids in many respects (see esp. Chapter 1), perhaps in
none more than architectural patronage. Beginning with the high-ranking
Samanid ghulam Alptigin (d. ah 366/977 ce) and his successor Sabuktigin
(d. ah 389/999 ce) – both considered progenitors of the Yamini–Ghaznavid
dynasty – all of this royal house’s scions had been reared in cosmopolitan
and courtly Perso-Islamic surroundings. Although of Turkic ethnic and
linguistic origins, these ghulams had been incorporated into the Samanid
court, witnessing its active encouragement of Arabic- and Persian-language
writing in both secular and religious fields.92 Characterizable at least ini-
tially as a Samanid “successor-state,” the Ghaznavid dynasty continued
to feel the benefits of maintaining Arabic-language bureaucracy, while
the rulers themselves were knowledgeable in and patronized literary pro-
duction in Persian (and also Arabic).93 This overall investment in a cul-
tural “footprint” further intertwined Zabulistan within the commercial,
intellectual, and religious networks of greater Eurasia (see also Chapters 2
and 3).
The Ghaznavids’ immersively courtly background could not form
a more glaring contrast to the Shansabanis’ obscure origins in seasonal
transhumance and/or nomadism. Nevertheless, the Firuzkuh Shansabanis
quickly learned the mechanisms of attaining and exercising contextually
appropriate kingship, including the commissioning of magnificent archi-
tectural complexes such as Shah-i Mashhad, Chisht, and the Firuzkuh
jami‘ complex and its renowned minar (the specific subjects of the present
chapter). Based on the significant differences between these sets of imperial
patrons, however, I suggest that the Shansabanis’ imposing buildings were
as much – if not more – the results of input from courtly intelligentsia and
the actual craftspeople who fabricated them, as from their royal masters.
For the Shansabanis, their transhumant/nomadic pre-imperial milieu
could have perhaps offered some opportunity to experience built forms.
This would have occurred as they encountered settled environments
along their seasonal routes of transhumance, which spanned the length
of historical Ghur. Although again a modern parallel, it is nonetheless
useful to note the visitations of migrating nomadic pastoralists to Sufi
khanaqahs, particularly when these are located along their paths of move-
ment and/or near their seasonal market areas.94 Regular contact with such

222 Iran to India
establishments, and indeed the nomadic visitors’ participation to varying

degrees in the khanaqah’s activities – for example, dhikr, urs, or other
gatherings – historically would have represented one of the principal
avenues of exposure to an array of Islamicate socio-religious practices,
and perhaps even the totality of some of these groups’ actual knowl-
edge of “Islam” and its associated architecture.95 It is thus conceivable
that, through their interactions with even minor architectural complexes,
nomadic/transhumant visitors would gain at least a cursory apprehension
of current styles, some among them also perceiving the buildings’ more
immediate functions and even their larger political implications.
Such regular seasonal, architectural, and urban interactions would have
contributed little, however, toward deepening knowledge of the Qur’an
and its exegetical traditions among groups such as the Shansabanis,
or giving them comprehensive training in Persian literary history and
composition.96 In fact, taking into account their limited familiarity with
Persianate courtly cosmopolitanism, it is understandable – rather than
remarkable – that virtually no Persian inscriptions appeared on the impe-
rial Shansabanis’ buildings; their predilections naturally tended toward a
select and effectively reduced group of Qur’anic verses and simple doxolo-
gies, which were by default in Arabic.97 The differences between the rulers
and at least some of their courtiers with regard to Qur’anic and exegetical
conversancy would have to be bridged, then, in the course of selecting
verses, Hadiths or other theologically significant material as part of devis-
ing architectural epigraphy. The religious and intellectual experts in the
Firuzkuh Shansabanis’ retinues and court would have conceivably made
the initial selections from appropriate sources, even if these were pending
the approval of their patron(s).
In the end, both in the historical moment and in hindsight, it accrues
to the Shansabanis’ credit that they understood and enacted, with great
responsiveness, the urgency of building monumentally for an emerging
transregional elite such as themselves: commissioning and endowing
“civic” institutions was an effective way of commanding and mobilizing
labor in their areas of control, even without erecting palaces. The actual
fruits of this expenditure of resources, namely, the prominent architectural
complexes centering and sustaining their environs, were not only attesta-
tions of economic wherewithal, they were also the most efficacious means
of monitoring and eventually controlling legal thought and practice,
which are among the very mechanics of empire along with military might
and commercial prosperity. In the case of the Firuzkuh Shansabanis, the
age-old Khurasani commercial networks and prosperous urban centers
remained part of their architectural–cultural habitus; it is all the more
noteworthy, then, that they apparently persisted in their preference of tent
over palace, continuing their seasonal migrations between their summer
and winter encampments or “capitals.”

Notes
1. The dating of Shah-i Mashhad is not definitive due to the erosion of its
inscriptional program, particularly the plaited Kufic on the principal south-
ern portal (Figures 4.1 and 4.4) (inscription Nos. 1 and 2 in Casimir and
Glatzer 1971). The initial reading of the date was ah 561/1165–1166 ce, modi-
fied later to ah 571/1176 ce, which continues to be accepted. Cf. Casimir and
Glatzer 1971: 56; Glatzer 1973: 50; Blair 1985: 81. Najimi (2015: 148 n. 33, 151
n. 54), while not attempting to re-read the date itself in the inscription, does
reconcile both proposed years by stating that a decade-long construction of
the complex was realistic, “based on his experiences working on restoration
projects in Afghanistan.”
2. Cf. Introduction esp. for the Salar Khalil, west of Balkh (Juzjan Province),
possibly falling within the ambit of the Bamiyan Shansabanis. These architec-
tural remnants are datable to the later twelfth century, likely of non-royal but
still elite patronage. See also Introduction and Chapter 1.
3. The Appendix gathers together the monumental historical and religious epig-
raphy from these structures, serving as a complementary reference to the
discussion in this chapter and the following one of the import of this inscrip-
tional corpus.
4. Blair 1985: 83.
5. Nishapur’s sectarian riots are well known, thanks to the work of R. Bulliet
and others: here, prior to the city’s demise in the wake of the Oghüz raids of
the 1150s, the two predominant and rival madhhabs of the Shafi‘is and Hanafis
were prone to violent confrontations. Among other Sunni orthodox sects,
the Hanbalis and Malikis were relatively few in number, mustering less vio-
lence-inciting fervor. The Karramiyya also, even if marginalized as heterodox
(discussed infra in main text), periodically amassed enough clout especially
among the poor in the northwest of the city, to instigate violent eruptions
against the dominant religious sects. Cf. Bulliet 1972: 12, 30, 74, 76; Zysow
2011/2012: pt ii; and Chapter 2.
6. For the association of Qur’anic inscriptions and Hadith with certain madh-
habs, see Blair 1992: esp. 9–10, 52–53, 78. For the minar of Jam-Firuzkuh, see
Sourdel-Thomine 2004: 153ff.; Flood 2005a: 270–273; and infra in main text.
7. While Blair (1985) first deciphered and published several of the Chisht inscrip-
tions principally in the more intact southern Structure A, Dr. Viola Allegranzi
is in the process of re-examining, transliterating, and translating the inscrip-
tions in both the structures surviving at the site (see below). Her work is based
on Josephine Powell’s photographs from the mid-twentieth century, as well
as a comparison with Ludvik Kalus’ readings of the inscriptions, published on
the digital database Thesaurus d’Epigraphie Islamique (TEI): cf. fiche 37804. I
am deeply grateful to Dr. Allegranzi for sharing her work in progress, which
henceforth is cited as Allegranzi 2019b (forthcoming). See also Ball 2008:
179–181; Ball 2019: No. 212 for a description of the site, plans and sections.
8. A possibly similar anachronism occurred on an undated jital minted at
Kurraman (modern Parachinar), between Peshawar and Gardiz, naming
one Shams al-Din as the sovereign. The coin has been attributed to the

224 Iran to India
Bamiyan ruler Shams al-Din Muhammad (r. ah 558–588/1163–1192 ce), since

the ruler of Firuzkuh had relinquished this laqab in favor of Ghiyath al-Din
by ah 588/1163 ce. Hence, the dilemma: the 1160s would be too early for the
Shansabanis to have a territorial presence this far east; and yet, it is equally
unlikely that the Bamiyan Shansabanis ever reached Gandhara, even at the
height of their power in the 1190s ce. Cf. Tye and Tye 1995: 53 and No. 138;
also O’Neal 2015. For a parallel instance in the case of Shihab al-Din/Sultan
Mu’izz al-Din, see Chapter 1 (notes).
9. Cf. Juzjani vol. I, 1963: 354 (trans. vol. I, 1881: 369). An examination of
Shansabani marital alliances could offer additional insight into the differing
politico-economic wherewithal and consequent hierarchy among the various
lineages. Over time and with Firuzkuh’s growing coffers – refilled via a trade
economy – the Firuzkuhi Shansabanis were able to enter marriage alliances
outside Ghur (see main text), which were also more useful than marriage
among kin. Meanwhile, the Bamiyan Shansabanis’ marriages appear to have
been largely among their Shansabani agnates, e.g., the marriage of Sultan
Shams al-Din Muhammad (r. 1163–1192), Fakhr al-Din Mas‘ud’s successor,
to Hurra-yi Jalali, the elder sister of the Firuzkuh and Ghazna sultans Ghiyath
al-Din and Mu’izz al-Din (respectively), and the mother of the great cultural
patron Baha’ al-Din of Bamiyan (r. 1192–1206; see main text). Additionally,
Malika-yi Jalali, Sultan Ghiyath al-Din’s daughter, was betrothed to Baha’
al-Din’s son and heir Malik ‘Ala’ al-Din Muhammad. Cf. Juzjani vol. I, 1963:
379, 388 (trans. vol. I, 1881: 412, 428). Barfield (1981: 78) noted that modern
nomads in northeastern Afghanistan rarely arranged matches among patrilin-
eal cousins, instead preferring outside alliances and the larger affine networks
afforded by such exogamous marriages. By contrast, paternal cousin marriage
continues in force among the Arabian Bedouin (Barfield 1993: 77), an old
practice that could have been the source of the orthodox Islamic prescriptions
for such arrangements.
10. Juzjani vol. I, 1963: 337, 341, 349 (trans. vol. I, 1881: 341). Sourdel-Thomine
(1976: 168) proposed that this site was the same as Shah-i Mashhad, citing
both Juzjani and Le Strange (1930: 416). However, the location does not coin-
cide with Shah-i Mashhad: cf. Glatzer 1973: 67, whose comments on the loca-
tion are particularly weighty given his rediscovery of the complex, together
with M. Casimir, in 1970 (Casimir and Glatzer 1971). See also Najimi 2015:
147; Wannell 2002: 242–243; and infra in main text. W. Ball (pers. comm.,
April 2020) has suggested that Shahr-i Arman could be the historical Afshin
– cf. Ball 2019: No. 1033.
11. Juzjani vol. I, 1963: 362–363 (trans. vol. I, 1881: 385). The prominent mention
of this individual may indicate that he was actually Muhammad ibn Haisam,
the founder of the eponymous Haisamiyya of the later twelfth century. Ibn
Haisam put forth significant revisions of Karrami theology and thus estab-
lished this important “sub-sect” among the Karramiyya, with whom Fakhr
al-Din Razi also engaged in his writings. The Haisamiyya were one of three
such sub-sects representing later developments in Karrami thought. Cf.
Zysow 2011/2012: pt vi.
12. For Taj al-Harir Jauhar Malik, see Juzjani vol. I, 1963: 357 (trans. vol. I, 1881:

376); also Casimir and Glatzer 1971: 56ff., Blair 1985: 81ff.; Najimi 2015: 149.
For all sources on the complex and site, see Ball 2019: No. 1023.
13. Najimi 2015; see also below.
14. Juzjani vol. I, 1963: 354 (trans. vol. I, 1881: 367). As mentioned in Chapter
1, it might have been during these military actions that Saif al-Din briefly
occupied Herat, where coins were issued with his laqab and the Herat mint
stamp (see also O’Neal 2016a; O’Neal 2020: 200–201).
15. In this case, the apparent deviation from the generally exogamous marriages
of the Firuzkuh sultans (note above) was in fact an extremely convenient
use of the patrilineal cousin marriage prescribed within Islam. The mar-
riage of Jauhar Malik and Shams al-Din presumably took place between
ah 556–558/1161–1163 ce: it must be remembered that ‘Ala’ al-Din Husain
Jahan-suz, upon acceding to the throne of Firuzkuh in ah 544/1149 ce, had
imprisoned both the sons of his brother Baha’ al-Din, likely when the boys
were quite young. After ‘Ala’ al-Din’s death in ah 556/1161 ce and the acces-
sion of his son Saif al-Din, the latter released his cousins and, perhaps by way
of healing the familial breach, married his sister to his elder cousin Shams
al-Din. Shortly after the nuptials, the unforeseen death of Saif al-Din occa-
sioned yet another transferal of the Firuzkuh throne to a collateral Shansabani
lineage (cf. Chapter 1). In hindsight, then, this fourth transferal was surely that
much smoother and more acceptable by virtue of the marriage of Saif al-Din’s
sister to the eventual Sultan Ghiyath al-Din. Cf. Juzjani vol. I, 1963: 346, 395
(trans. vol. I, 1881: 357, 446–447).
16. Although documented in the 1970s and again in the 1990s, the modern inter-
actions of nomadic populations with sites such as Shah-i Mashhad – precisely
what led to the site’s “discovery” by Casimir and Glatzer (1971) – may provide
a useful parallel to the historical contacts between these groups and the mon-
umental complexes dotting their migration routes. The notable deterioration
in the complex within twenty years (Figures 4.1 and 4.4), however, is not
necessarily attributable to this interaction, but rather to the agriculturalists of
the region, who had greater need for the building materials readily available
at abandoned monumental sites.
17. Franke and Urban and their archaeological team documented structures like
Gunbad-i Shohada and Qala-yi Sarkari, both between Awbeh and Chisht,
and Rabat-i Chauni, south of Chisht. Gunbad-i Shohada could have been a
commemorative structure of the recognizable “domed cube” typology, quite
plausibly attributable to late Saljuq/Shansabani presence (or later); the Qala
and Rabat are square-walled compounds with corner circular towers, which,
as we have seen in the case of Danestama (cf. Chapter 3), could have been
pre-Islamic structures that were reused or repurposed during later periods,
to be determined only with further study. Cf. Franke and Urban 2006: 7–15,
18–20, 22–23; and infra in main text.
18. Flood 2009a: 101; see also Chapter 2.
19. Sourdel-Thomine (2004: esp. 156ff.) first proposed the reinterpretation of
the minar’s inscriptional program. See also Flood 2005a: 269–273; and Flood
2009a: 96–97. The Jam-Firuzkuh minar was also discussed in Chapter 3 – see
also infra in main text.

226 Iran to India
20. Juzjani vol. I, 1963: 362 (trans. vol. I, 1881: 384). For the Karrami Qur’an,
see Flood 2009a: 96; Flood 2009b. Thomas (2018: 122) briefly discussed the
overall complex; as always, Ball (2019: No. 1023) has provided the most thor-
ough bibliography of the site.
21. For the Ghaznavids, see, e.g., Bulliet 1972: 63–64, 68–70; Malamud 1994:
45–46. For Saljuq-period negotiations of sectarian rivalries, see Bulliet 1972:
72–74; Peacock 2010: 104ff.; Peacock 2015: 268–272.
22. See below in main text for possibly indirect infiltrations of the orthodox
madhhabs into more remote areas, possibly via Sufi networks. Given the
founder Muhammad ibn Karram’s proselytizing origins in Sistan – though
he might have been born in Mecca – it is possible that he made his way from
Sistan to Khurasan via Ghur and Gharjistan, reportedly entering the great
Khurasani emporium of Nishapur with a following of “weavers and others
from depressed classes.” See esp. Bosworth 1973: 165, 185–186; Melchert 2001;
also Zysow 2011/2012: pt i.
23. Bosworth 1973: 185–186.
24. See also Bosworth [1960] 1977 (reprint): 1–2.
25. Bosworth (1961: 128ff.) initially put forward the idea that Mahmud dis-
patched Karrami preceptors into Ghur, but see also Malamud 1994: 47;
Zysow 2011/2012: pt ii; and Chapter 3. These later Karrami “missionaries,”
then, quite possibly entered the region with such information as might have
been preserved ever since a century before or more.
26. Cf. Bosworth 1973: 188–189; Malamud 1994: 39–43 and notes.
27. See Bosworth [1960] 1977: 8; also Zysow (2011/2012: pt iv) for a discussion of
early concordances among Karramis, Shafi‘is and Hanafis, as well as eventual
divergences between the Karramiyya and other madhhabs. It is also impor-
tant to bear in mind that, prior to the thirteenth century, as many as several
hundred lesser known or lost theological sects were active throughout the
central Islamic lands, and that the four madhhabs now accepted as “orthodox”
crystallized and came to be considered as such only by the later 1200s. Cf.
Makdisi 1971: 77; Makdisi 1981: 2ff.
28. Cf. Bosworth 1973: 183. According to Nizami (1998: 370–371), Ghur and
Gharjistan were already crowded fields for proselytizers by the early twelfth
century: in addition to the Karramiyya, disciples of the mystic ‘Abd al-Qadir
Gilani (d. ah 561/1166 ce) were also active in these same areas.
29. Recently, the minar of Jam-Firuzkuh has not only been re-dated from ah
590/1193–1194 ce to ah 570/1174–1175 ce, it has also been compellingly iden-
tified as a Karramiyya-inspired monument, particularly its unique epigraphic
program (Sourdel-Thomine 2004: 156; cf. further citations below in main
text and notes). The architectural typology of the minar versus the madrasa,
however, sets the structures apart, as discussed infra (main text).
30. See esp. Tritton 1957: 98–100, 102–108; Sourdel-Thomine 1976; Pedersen,
pt i and Hillenbrand pt iii in Pedersen et al. 2012; Zysow 2011/2012: pt ii.
Melchert (2001: 237) in fact cited al-Maqdisi’s (fl. 985) characterization of the
Karramiyya as al-khanaqa’iyyun. It is noteworthy that, well into the twelfth
century, textual sources also included ribats as sites of instruction, not only in
smaller towns but also in larger cities such as Baghdad; at times a communal

structure previously at the edge of a shifting frontier – the very definition

of ribat – would be essentially repurposed as a place “for those devoted to
the contemplative life … [and] for the spreading of learning.” Cf. Tritton
1957: 106–108 and notes; Makdisi 1981: 9–34; Böwering and Melvin-Koushki
[2010] 2012: pts i and ii; Edwards 2015: 120–121.
31. One of the methods of implementing imperial Saljuq control was via the
patronage of various Sunni madhhabs in major urban centers – patronage that
included the construction and endowment of lavish madrasas. See Böwering
and Melvin-Koushki [2010] 2012: pt ii. Bombaci (1966, cited in Allegranzi
2019a, vol. I: 161 n. 148) proposed that a possible precedent for the imposing
Saljuq madrasas, promulgated particularly by the famous vizir Nizam al-Mulk
(ah 409–485/1018–1092 ce), might have been one of Ghaznavid patronage in
Nishapur, dating to c. 1000 ce. See also Bulliet 1972: 250–251. For an acceler-
ated proliferation of madrasas due to Saljuq patronage, see Makdisi 1971: 81–82;
Makdisi 1981: 31ff.; Bulliet 1994: 146–147 and passim; Peacock 2015: 211–215.
32. The classic studies by Bulliet (1972, 1994) on sources for Nishapur during the
eleventh–early twelfth centuries demonstrate the intimate intertwining of
economic, political, and intellectual interests among the “patricians” of the
city. Nonetheless, there were rifts and reconciliations among them, and also
with the rulers, depending on the situation – cf. also Bosworth 1973: 192–193.
33. Well into the eleventh century, while the Karramiyya were still in Ghaznavid
favor in Nishapur and had attained high positions even in non-religious
administration of the city, asceticism continued to be associated with them.
Cf. Bulliet 1972: 42, 64ff.; esp. Bosworth 1973: 187–188; Bosworth [1960]
1977: 6–7. Whether this asceticism was “hypocritical” (Bosworth [1960] 1977:
11) or not, the rapid decline in the Karramiyya’s fortunes shortly thereafter,
and the consequent shift of their proselytism into the hinterlands, would have
required even more that the adherents maintain frugal if not ascetic lifestyles.
34. By the tenth century, there were madrasas in Egypt that were occupied
by at least two madhhabs, wherein instruction according to each one took
place at different locales throughout the complex, or if space was lacking,
timings would be split. The multi-madhhab institution was not uncommon
through the fourteenth century, and included such prominent foundations as
Baghdad’s Mustansiriyya, founded in ah 631/1234 ce to contain “four schools
for the four rites …,” effectively eclipsing the Shafi‘i NiÕamiyya (founded ah
459/1067 ce). Cf. Tritton 1957: 100–101, 104–105; and esp. Makdisi 1981: 34.
35. Blair 1985: 81.
36. I am grateful to Gwendolyn Kristy and David Thomas for helping determine
Shah-i Mashhad’s orientation: Gwendolyn Kristy located ArcGIS imagery
of the site, and David Thomas superimposed on it the complex’s plan as
reconstructed by Najimi (2015: 153), whose own plan ensued from the studies
by Casimir and Glatzer (1971: 55; cf. also ibid., 58) and his own documenta-
tion of the madrasa’s standing remains in the early 1990s. The accurate qibla
orientation is also from Najimi (2015: 148).
37. Cf. Akhmedov 2000: 195–199. Berggren (2000: 189–190) noted that “[the]
mathematical sciences acquired Islamic dimensions as its practitioners
became aware that their disciplines could be used to provide exact solutions to

228 Iran to India
problems unique to Islamic societies … Trigonometry found several areas of

application, one of these being the determination of the direction of prayer,
i.e. the determination of the direction of Mecca (the qibla), for a given local-
ity.” But, cf. infra in main text.
38. The astronomical thinking of Hanafis and Shafi‘is in Samarqand during the
eleventh century is known thanks to a text by al-Bazdawi (d. ah 482/1089 ce,
of which a seventeenth-century manuscript survives in Cairo): the Shafi‘is
simply followed the precedent of the Prophet’s qibla in Medina, which
faced south – coincidentally also an auspicious direction in pre-Islamic wind
theory, whence the south wind qabul (sharing the Arabic root q-b-l with
qibla) brought rain to the Hijaz. The Hanafis, by contrast, relied on their
immediate surroundings as determinants, basing their western qibla on the
direction of the road from Samarqand to Mecca. See King 1982; King 1995;
King 1993.
39. It should be observed that, given Karrami leanings toward Hanafi legal tenets
(cf. Zysow 2011/2012: pt v), it is likely that the Karramiyya also adhered to
a qibla due west (270°). The likelihood is bolstered by the minar of Jam, a
monument in all probability of Karrami affiliation, which has a mihrab on its
east face oriented 260°, very close to the Hanafis’ western qibla (“true” qibla
at Jam would be 245°). D. Thomas, pers. comm., May 2018; see also infra in
main text and Chapter 3.
40. King 1992: 254.
41. See Najimi 2015: 154. It is lamentable that the author did not include a
photograph of the previously unpublished inscription to aid in seeing its
exact location within the epigraphic program of the southern façade, its scale,
and its relationship with the façade’s other inscriptions. Cf. also Appendix,
Afghanistan IV:2.
42. Najimi 2015: 158 (emphasis added); and ibid., 167–168 for a table listing the
varying readings of some of the inscriptions. See also Casimir and Glatzer
1971: 56ff.; Glatzer 1972: 50ff.; Blair 1985: 83–86.
43. Casimir and Glatzer 1971.
44. Sourdel-Thomine (2004: 153–154) described in some detail the complexity
involved in planning the monumental epigraphic program on the minar of
Jam (cf. Chapter 3 and infra in main text), but also applicable to other sites.
The challenges to overcome included the varying lengths of verses, and their
placement such that size and script modifications according to space would be
virtually unnoticeable. These factors were relevant also in the opening verses
of Surat al-Fath: significant differences in length among verses 5–8 make it
difficult to surmise how many would have continued on the no longer extant
southwestern flank of the façade. Cf. also Ettinghausen 1974: 307–309.
45. Najimi 2015.
46. Tashriq refers to the eleventh through the thirteenth days of the month of
Dhu al-Hijja, three days after ‘Id al-Adha and also the three-day period of
the end of the Hajj. It is the time of animal sacrifice, which in essence hapti-
cally recalls Abraham’s (Ibrahim) willingness to sacrifice his son upon divine
command. However, there was apparently no specific takbir to be recited
during this period, at least according to al-Bukhari’s Sahih. Cf. at: https://

sunnah.com/bukhari/25 (Sahih Book 25, “Kitab al-Hajj”) and https://sunnah.

com/bukhari/30: (Sahih Book 25, “Kitab al-Saum”). See also Appendix,
Afghanistan IV: 3.
47. According to Najimi 2015: 163 (Appendix, Afghanistan IV: 9); but see infra in
main text.
48. Blair (1998: 69) claimed that verses from Surat al-Fath were common on
mosques, though evidently this assessment applied outside the eastern Islamic
lands: Blair (1992) did not document a single occurrence on any building
type through ah 500/1106 ce. However, at Barsiyan within the Isfahan oasis,
Q. XLVIII: 1–5 appeared at the base of the dome of the complex’s mosque
– a complex including a minar and possibly a caravanserai; the mosque was
dated by inscription to ah 528/1134 ce (about thirty-six years after the minar’s
inscribed date of ah 491/1097–1098 ce) (cf. Godard and Smith 1937: 40–41).
Finally, late eleventh- and twelfth-century quotations of the Sura’s verses (Q.
XLVIII: 1–5) come from Cairo, namely, at Mashhad al-Juyushi dating to ah
477/1085 ce; and a century later at the adjacent, “magnificent Citadel” (Ar.
qal‘a) of Cairo (Q. XLVIII: 1–3), initiated by the renowned Salah al-Din in
ah 578/1183–1184 ce. Cf. RCEA, vol. 9 (1937) No. 3380; and Rabbat 1995:
68–73. Upon this first citation of the RCEA, we should bear in mind early
critiques of this monumental project even by one of its editors, J. Sourdel-
Thomine (1976: 265), who questioned “the role to be played … of a tool
[whose] conception was already old [i.e., obsolete].” Hillenbrand (2012: 19)
specified the RCEA’s in-built obsolescence and its consequences: “Strange as
it may seem, Quranic inscriptions are recorded in this vast opus only when,
as part of the same inscription, historical matter is included.” Such is the
case with the aforementioned Barsiyan complex, whose dome and mihrab
inscriptions were not noted in RCEA – despite the latter’s mention of a date
(cf. also Kalus, TEI fiche No. 37215; Godard and Smith cited supra). It is
possible, then, that important instances of Qur’anic citation may not factor
into scholarly analyses simply because they do not appear in the RCEA. Cf.
also Appendix.
49. All identified Shansabani-patronized epigraphy is in Arabic, with the excep-
tion of the Persian date of Chisht’s southwestern structure (discussed infra
in main text; Appendix, Afghanistan III: 5d). This overall trend is at vari-
ance with Ghaznavid and Qarakhanid epigraphic corpora, wherein Persian
was increasingly used particularly as a language of architectural inscriptions,
beginning in the early eleventh century and onward. Cf. Hillenbrand 2012:
21ff.; esp. Allegranzi 2015: 34 and passim; Allegranzi 2019a, vol. I: 47ff.
50. The actual readability of monumental epigraphy in pre-modern times has long
been discussed in scholarship, beginning with Ettinghausen (1974). Scholarly
consensus has stressed that “Islamic inscriptions were not expected to be read
in this literal sense but rather functioned in an imagistic or symbolic manner,
affirming the belief of the patron as well as that of the viewer” (Edwards 1991:
65). See also Thomas 2012: 145–146, 156–158.
51. It is more than likely that, well into the mid-twelfth century, the Firuzkuh
Shansabanis also maintained some aspects of non-Islamic religious practices.
Cf. Chapter 3 for Tiginabad/Old Qandahar in the plains of al-Rukkhaj,

230 Iran to India
where their forays in ah 552/1158 ce – only fifteen years before the conquest of
Ghazna – seem to have left sparse but distinctive traces, e.g., in burial mounds:
“hybrid” burial practices combining a qibla orientation of the deceased with
unusual underground vaults and octagonal platforms above ground. Indeed,
the south- and eastward campaigning armies of Mu‘izz al-Din would have
continued to encounter varieties not only in tomb structures, but also in
burial practices, e.g., in Baluchistan and coastal Makran, where they cam-
paigned in the 1180s (cf. Chapter 6). For these contiguous areas, cf., e.g.,
Hassan 1991: 79ff.
52. Khazeni’s (2009: 13 and passim) work on central Iran’s Bakhtyari tribes within
Qajar realms (late nineteenth–early twentieth centuries) described its most
useful source, Tarikh-i Bakhtyari, as “a tribal history and perhaps the first
ethnography in the Persian language,” compiled between 1909 and 1911; it
was a sweeping combination of relevant passages from official Persian chron-
icles and “original material … devoted to the oral histories and geographical
lore of the tribes, including … organization, administration, and customs.”
Historical empires originating in nomadic pastoralism, such as the Qara-
or Ilak-khanids (mid-tenth–early thirteenth centuries) and the Qarakhitai
(1087–1143; cf. Chapter 1), evince divergent literary developments strongly
impacted by contact with the Islamic-Persianate ecumene: the Qara- or Ilak-
khanid Satoq Bughra Khan Abd al-Karim’s (d. c. 955 ce) embrace of Islam
led not only to a wave of conversion among other Turks, arguably it also led
to the creation of a Turkic – rather than Persian – literary tradition begin-
ning in the 1070s. Cf. Biran 2012; Vàsàry 2015: 17ff. By contrast, modern
scholars of the unconverted Qarakhitai still come across great difficulties
when searching for primary sources, not least because of “the irregular record
keeping of the nomadic Khitans and the unusually long time that passed
from … 1125 to the compilation of the [Chinese Liao shi] in 1344–45” (Biran
2005: 4).
53. For the larger juridical and even theological debates in which the Karramiyya
engaged with other madhhabs, see Zysow 1988: 585 and passim; Zysow
2011/2012: pts iv and v.
54. Cf. Edwards 1991: 69.
55. Cf. Najimi 2015: 158–163 for a detailed description of the manufacture of
inscriptions in brick and stucco; see also Chapters 2 and 3, for a discussion
of the availability of local practitioners of the architectural culture prevalent
throughout Khurasan. In the context of the Jam-Firuzkuh minar – cf. infra in
main text – Lintz (2013: 96–97) also discussed collaboration among builders,
artisans, theologians, and patrons.
56. For the typical career of a purchased Turk slave, see Barthold (1968: 227–228),
who derived the trajectory from the Saljuq vizir Nizam al-Mulk’s Siyasat-
nama. For the Ghaznavids in particular, Allegranzi (2014: 113–114) summa-
rized Baihaqi’s praise of Mas‘ud I’s skills as architect/engineer (muhandis)
(also Bosworth 1973: 139–141). Although this could have been rhetorical flour-
ish in praise of his patron, it nonetheless appears that Sultan Mahmud had
been diligent about the education of his sons in Persian and Arabic adab,
and the early Ghaznavid sultans’ courts “[u]ndoubtedly … became brilliant

cultural centers” (ibid., 131ff.). Cf. also Allegranzi 2015: 24–27. For a summary
of the extensive patronage of poetry throughout the length of the Ghaznavid
dynasty, see Allegranzi 2019a, vol. I: 23–31. See also Appendix, Afghanistan V
and VI.
57. For Egypt and the Mediterranean, see, e.g., RCEA, vol. 5 (1934) Nos. 1838,
1879, 1917, 1931, 1938, 1950. In the eastern Islamic lands, two tomb towers
at Kharraqan dating to the late eleventh century ce carry Q. LIX: 21–24
(Blair 1992: 9, 134, 172). According to Dodd and Khairallah (1981: II:130–131),
whose work concentrated on architecture rather than smaller elements (e.g.,
epitaphs), parts of this verse series was first documented in Córdoba’s jami
dating to ah 354/965 ce, and thereafter principally in India, beginning with
Delhi’s Qutb mosque in ah 587/1191–1192 ce and continuing through the
fifteenth century (see Volume II).
58. In Blair’s (1992) compilation of inscriptions from greater Iran and Transoxiana
of the ninth through early twelfth centuries, no takbir was documented, and
only one Hadith – on a mihrab from Iskodar, dated to ah 400/1010 ce. Cf.
Blair 1992: 10, 78; also Blair 1998: 69. At variance with Blair, Hillenbrand
(2012: 19) claimed that Hadiths “often appear[ed] in building inscriptions.”
In fact, Hadith may have gone unrecorded in the RCEA because, unlike
Qur’anic inscriptions, Hadith were not as often accompanied by historical
data (date of foundation, patron, etc.) – the collection of which was the true
aim of the monumental RCEA. See also supra in notes.
59. For Shi‘a presence in Khurasan, cf. Bulliet 1972: 14–15; Bosworth 1973: 194–
200; Bosworth 2012c. It is also noteworthy that ‘Ali’s name was repeated (in
Kufic) on the eastern pier of the madrasa’s northern entrance, adjacent to
the Hadith under discussion (cf. Casimir and Glatzer 1971: fig.12; Najimi
2015: 167). Moreover, in the epigraphic programs of the known Shansabani-
patronized structures throughout modern Afghanistan and Pakistan, ‘Ali
was included in invocations of the Rightly Guided Caliphs of early Islam,
e.g., in the stucco medallions decorating the mihrab of Chisht’s northeast-
ern structure, and in similar brick medallions on the mihrab of the ribat
of ‘Ali ibn Karmakh near Multan. For Chisht, see Allegranzi 2019b (forth-
coming); for the Multan ribat, see Edwards 1991: 92; Edwards 2015: 110–133,
201–210; and Chapter 6. See also Appendix, Afghanistan III: 7b; and Pakistan
I: 7.
60. Cf. at: https://sunnah.com/bukhari/81, or Sahih al-Bukhari 6416, Book 81
(print Vol. 8, Book 76, Hadith 425). See also Melchert 2012; and Appendix,
Afghanistan IV: 9.
61. Pinder-Wilson 2001: 164ff.; cf. also Appendix, Afghanistan VI. Prior occur-
rences of Surat al-Fath’s opening verses occurred at Qasr Kharana, Jordan (Q.
XLVIII: 2) in ah 92/710 ce, and in the cupola of Cairo’s al-Juyushi Mosque
(Q. XLVIII: 1–5) in ah 478(?)/1085(?) ce. Farther east in the Iranian world,
the verses (Q. XLVIII: 1–5) appeared in the Isfahan oasis’ Barsiyan mosque.
Cf. Godard and Smith 1937: 40–41; Dodd and Khairallah 1981: II: 118–120;
and supra in notes.
62. The minar also survived the destruction left in the path of the Mongol contin-
gent in pursuit of the Khwarazm-shah Jalal al-Din Mingburnu in the 1220s,

232 Iran to India
and the succeeding several centuries when the descendants of the Mongols
– e.g., the Negüderi, or Qaraunas – held sway from Ghazna eastward. Cf.
Bernardini 2020: 251–252; and infra in this chapter.
63. Pinder-Wilson 2001: 165. See esp. Bosworth [1977] 1992: 6ff. This “sequel”
to the same author’s earlier publication on the Ghaznavids (1973) essentially
began with the Saljuq defeat of the Ghaznavid forces at Dandanqan, which
arguably brought to an end the expansionist momentum set by Mahmud and
continued by Mas‘ud I. During the following eighteen years, thanks to heavy-
handed Saljuq meddling in Ghaznavid succession, six Ghaznavid princes
vied for the title of sultan either as claimants or actual rulers. Thereafter,
the combined reigns of Ibrahim (r. ah 451–492/1059–1099 ce) and his son
Mas‘ud III (r. ah 492–509/1099–1115 ce) – ostensibly a long and seemingly
stable period of more than a half-century – again resulted in short and con-
tested successions (see also de Bruijn 1983: 34, 59). As noted above in the main
text, the Shansabanis sacked Ghazna in ah 545/1150 ce. It is probable that the
Oghüz occupation of Ghazna began as early as 1160, since Khusrau Malik
(r. ah 555–582/1160–1186 ce) was apparently crowned sultan at Lahore. Cf.
Bosworth [1977] 1992: 82ff., 123–131.
64. See Pinder-Wilson 2001: 166; Ball and Fischer 2019: 477–478. Mas‘ud III’s
minar was most likely associated with an adjoining mosque, and also threw
its shadow on the large Ghaznavid palace (R. Giunta, pers. comm., July
2018), discussed in Chapter 5. The precedent of commemorating a military
victory with the erection of a minar – in association with a mosque – appears
to have been firmly set already by Mahmud of Ghazna, who, according
to Iltutmish’s (r. ah 618–633/1221–1236 ce at Delhi) court poet Fakhr-i
Mudabbir Mubarakshah (d. ah 625/1228 ce), commemorated what was likely
his Kanauj victory of ah 409/1018–1019 ce by building a minar in Lahore (see
Flood 2002: 103–107, also for the poet’s mistake in the date of the campaign).
Due to Lahore’s millennium-long urban development, neither the minar
nor its surrounding context is traceable. Based on surviving towers – e.g., at
Ghazna, Jam-Firuzkuh, and Delhi – it is probable that the Lahore minar was
associated with a mosque at some point, though the latter may not have been
constructed simultaneously, as was the case at Jam-Firuzkuh and Delhi (cf.
also A. N. Khan 1988: 304; Chapters 3 and 5; and Volume II for Delhi).
65. Evidently the Sura in its entirety appeared again in Isfahan, around the court-
yard of the Imami madrasa, in ah 741(?)/1340(?) ce. See Dodd and Khairallah
1981: II:118; and supra in notes.
66. Cf. Y. Godard 1936: 367–369; Godard and Smith 1937: 351; Pinder-Wilson
2001: 158.
67. Cf. Edwards 1991: 64.
68. Blair 1985: 83. The madrasa’s funerary aspect – probably memorializing Sultan
Saif al-Din (r. ah 556–558/1161–1163 ce), as noted supra in main text – would
have very aptly combined with commemorating the fateful military victory
on the heels of which he was assassinated.
69. Again, Thomas (2018: 122–123) briefly addressed the complex, but inaccurately
states that it was “sponsored by Ghiyath al-Din” (cf. infra in main text). Ball
(2019: No. 212) considered the two structures together, the western one being

a madrasa and the eastern one a mosque (cf. Figure 4.16). T. Lorain (Director
of the Mission Archéologique Franco-Afghane – Bamiyan; MAFAB) has doc-
umented another example of two proximate domed structures indicating a
larger original complex – possibly a madrasa – in the Bamiyan basin’s Fuladi
Valley. The site is now known as Khwaja Sabz Posh and has been attributed
to the period of Ghaznavid or Shansabani dominance of the region. T. Lorain
cited in Allegranzi 2020, n. xvii.
70. Cf. Blair 1985: 81–82; Allegranzi 2019b (forthcoming); Appendix, Afghanistan
III: 4.
71. Cf. Appendix, Afghanistan III: 5d. Blair’s (1985: 82) reading of the date was
slightly corrected by Giunta (2010b: 177). The combination of Arabic and
Persian in the same inscription was not uncommon in the epigraphy of
the eastern Islamic lands during the eleventh through thirteenth centuries.
Chisht’s inscriptional program continued within the regional tendency of
inscribing funerary and construction texts (and of course Quranic verses) in
Arabic, while poetry and at times also dates were in Persian. Cf. Allegranzi
2015: 35; Allegranzi 2020, forthcoming.
72. Cf. Allegranzi 2019b (forthcoming); and Volume II.
73. I am grateful to Dr. Viola Allegranzi (pers. comm., September 2019) for
sharing her ideas, to be published in a forthcoming work. See also Appendix,
Afghanistan III: 4–5. Juzjani (vol. I, 1963: 354; trans. vol. I, 1881: 370) simply
noted that Shams al-Din adopted the new laqab of Ghiyath al-Din, right
after Saif al-Din’s death and his own elevation to sultan and head of the
Shansabani confederacy (i.e., eldest male descendant of ‘Izz al-Din and his
legal wife, rather than the Turkic maid – cf. Chapter 3). This would have
occurred, then, in about ah 558/1163 ce. No doubt due to customary rever-
ential respect toward overlords and patrons, Juzjani himself anachronistically
referred to the Shansabanis with their regnal names even when narrating
events that occurred prior to official entitlement, e.g., at the births of both
Ghiyath al-Din and his younger brother Mu‘izz al-Din (vol. I, 1963: 353ff.;
trans. vol. I, 1881: 368ff.). However, the reverse anachronism at Chisht – of
using an outdated laqab years after a regnal one was adopted – is unusual
(Allegranzi, pers. comm., March and September 2019).
74. I am again grateful to Gwendolyn Kristy for obtaining satellite imagery of the
structures at Chisht and sharing it with David Thomas and myself; this gener-
ated extremely informative discussions among us, which to me demonstrated
the true intellectual generosity of these scholars.
75. Cf. Allegranzi 2019b (forthcoming); Appendix, Afghanistan III: 6–10.
76. While the “domed cube” form for tombs was in use by the tenth century
in Transoxiana (cf., e.g., Grabar 1966: 17), apparently not all tombs had
mihrabs. Tombs with mihrabs began to appear in the same period in the
eastern Islamic lands: cf. ibid., 21–22, 24, 32, 40; Patel 2015; and, of course,
the discussion of Baba Khatim in the Introduction. See also Chapter 6.
77. See Blair 1992: 9, who also noted that the verses appeared frequently in writ-
ings of theologians with Mu‘tazalite leanings, “as an allusion [to] and even
declaration of their dogma.” Cf. also Giunta (2010b: 124), who observed that
these verses were “since the age of Mahmud [of Ghazna], the most frequently

234 Iran to India
attested verses in funerary inscriptions (all on marble) of [Ghazna] during the

Ghaznavid and Ghurid eras.”
78. Cf. RCEA, vol. 5 (1934) Nos. 1652, 1824, 1831, 1837, 1881, 1896; RCEA, vol.
6 (1935) Nos. 2052, 2058, 2061, 2094, 2119, 2135, 2143,2145, 2183, 2190, 2194,
2196, 2200, 2205, 2206, 2211, 2218, 2315, 2327, 2332, 2343, 2350, 2351, 2354,
2361, 2363, 2371, 2372, 2383, 2399. See also Appendix, Afghanistan III: 5d. It is
noteworthy that verses falling within this range of Surat al-Iklhas (Q. CXII)
were prominently quoted both on the renovations I have assigned to Mu‘izz
al-Din at Lashkari Bazar, and on the tomb attributed to Ahmad Kabir in the
lower Indus region – cf. Appendix, Afghanistan VIII: 4; Pakistan III: 6, 20,
22; see also Chapters 5 and 6.
79. For a possible interaction between the Karramiyya and emergent Sufi
orders during the eleventh century – specifically the commonality in khan-
qahs between the two broad groups – see Trimingham 1971: 6–7; and esp.
Böwering and Melvin-Koushki [2010] 2012: phase I. Modern and contempo-
rary documentations of transhumant and nomadic communities have taken
note of their interactions with spiritual centers or khanaqahs in the course of
their migrations: e.g., Khazeni (2012: 137, 147–148) noted the eighteenth- and
nineteenth-century visitations of shrines by Türkman nomads; Utas 1980: 64;
Amiri 2020: 152–155. See also infra in main text.
80. Cf. Nizami 1998: 369. Precisely because of the described variety of origins
and practices among Sufi masters as well as devotees (see supra in main text),
a vast vocabulary accreted over time to refer to them: see esp. Ernst 1992:
5, 11; Green 2012: 8. Although removed in time from the twelfth century,
Anooshahr’s (2017) study of the Sufi Sheikh Muhammad Gauth Gwaliyari
(fl. sixteenth century) of the Shattari order provides a fascinating example
of the dissemination of a Sufi tariqa, in this case from early Safavid Iran to
Mughal India, the shift in political engagement of the order over time and
space, and some of the truly idiosyncratic rituals prescribed for gaining polit-
ical power. For the Shattari silsila, see also Nizami 1998: 378–379.
81. See, inter alia, esp. Nizami 1961: 182–184; Nizami 1998: 377–378; Böwering
[1991] 2011; Ernst and Lawrence 2002: 19ff; Auer, EI3 (2016).
82. The lack of association between the monumental Chisht remains and the
Chishtiyya would be further underscored by their purported adherence to
poverty and eschewal of political connections, which presented a stark con-
trast to the Suhrawardiyya, for example. Cf. Nizami 1961: 177–180, 240–243;
Trimingham 1971: 65–66; and Chapter 6. After Diez in 1924, Maricq and
Wiet (1959: 69–70) were the second European scholars to analyze (however
briefly) Chisht’s architectural ruins and their inscriptions, but connecting
the Chishtiyya only to the town rather than the remains of the monumen-
tal complex. Sourdel-Thomine (2004: 155) apparently drew the first direct
association between the coalescing Sufi order and the ruined complex. See
also Flood 2009a: 96–102. Nevertheless, this circumstantially derived and,
in the end, most likely inaccurate association was already operative as of the
later 1970s, e.g., in Utas’ semi-anthropological fieldwork on Sufi orders in
Afghanistan (Utas 1980: 65). Although circumspect, Blair (1985: 81–84) also
implied an association of the Chishtiyya with the two surviving structures at

Chisht, describing it as the site “where the founder of the Chistiyya [sic] order
of Sufis, Khwaja Abu Ishaq of Syria, had settled.”
83. See esp. A. Knysh’s (2017: 10ff.) insistence on this hyphenated usage in his
dense and engaged recent history of Sufism.
84. As recently argued, particularly for the Suhrawardiyya, by Khan (2016: 3
and passim), who considered “the Shi‘a milieu [as] probably the single most
important factor that facilitated the rise of tariqa Sufism.” Khan adhered to
Trimingham’s (1971: ch. I) tripartite developmental trajectory for Sufism
generally throughout the Islamic world, with tariqa Sufism falling within
1100–1600 ce. See also Chapter 5.
85. Bulliet 1972: 41–43. The Shafi‘is were not the sole madhhab to command
affiliation, however distant, from spiritual masters perceived by some as Sufis.
Indeed, texts by the famed Khwaja ‘Abd Allah Ansari (ah 396–481/1006–1089
ce) of Herat, “one of the outstanding figures in Khurasan in the 5th/11th
century,” not only emerged from a Hanbali background, Khwaja’s life overall
demonstrates the true complexity of Sufism. The term “Sufi” was fluid,
applied not only to renunciant masters – whose khanqahs were sometimes
infiltrated by truly antinomian qalandars (cf. Karamustafa 1994: 3–4) – but
also to Hadith scholars and polemicists, at times varying over one lifetime. Cf.
de Laugier de Beaureceuil [1982] 2011. See also Knysh (2017: 71–72ff.) for the
“early Muslims … subsequently co-opted into Sufism by its later proponents
…”
86. Sourdel-Thomine 2004.
87. Appendix, Afghanistan VII: 2, below. Cf. Ball 2008: 214–216 esp. for images
and diagrams of the minar. Upon first analyzing the minar’s epigraphy,
G. Wiet (1959: 27, nn. 1–2) has already noted that the recognizable phrase
“help from Allah and imminent victory” (“… nasr min Allah wa fath qarib”)
had been common as of the later tenth century in Fatimid Ifriqiya (cf. also
Sourdel-Thomine 2004: 128–129). Extricated from the remainder of the verse,
the phrase seemed to take on a doxological function, appearing particularly
on elite objects such as tiraz and ivory caskets – see, e.g., RCEA, vol. V (1934)
Nos. 1622, 1811, 1841, 1846, 1847, 1848, 1849. It also appeared in architectural
contexts, such as a stone construction text from Nablus, ah 410/1020 ce
(RCEA, vol. VI (1935) No. 2310); and a marble restoration text at Cairo’s Ibn
Tulun mosque, dating to ah 469/1077 ce (RCEA, vol. VII (1936) No. 2716).
Thereafter in the last decade of the twelfth century, Dodd and Khairallah
(1981: II:132) listed its appearance at Delhi’s Qutb Mosque in ah 587/1191 ce;
see also Husain 1936: 109; and Volume II.
88. Juzjani vol. I, 1963: 357–358, 396 (trans. vol. I, 1881: 376, 449); see esp. Sourdel-
Thomine 2004: 138ff.; Flood 2009a: 98; Flood 2009b: 93–94.
89. See Bulliet 1972: 203; Bosworth 1977; Malamud 1994: 39.
90. Cf. Bombaci 1966: 5ff.; Allegranzi 2015: 25ff.; Allegranzi 2019b (forthcoming);
Allegranzi 2020; Appendix, Afghanistan V: esp. 3–4; and infra in main text.
91. Blair 1985: 84. Further and more specific parallels were drawn between the
two architectural corpora by Pinder-Wilson 2001: 169–170. See also (inter
alia) Ball 2020 (forthcoming).
92. Cf. Meisami 1999: 16–17. Although referring to the Mu‘izzi ghulams active

236 Iran to India
in the north Indian plains in the 1190s, Siddiqui’s (2006: 14–15) observations
are applicable to the higher rungs of the Turk slave market: “[T]hey were
Persianized generals having nothing in common with the freeborn Saljuq or
Ghuzz Turks … They were brought up as Muslims and trained in Muslim
cultures and manners; the education of a slave was a good investment because
an educated and cultured slave fetched a higher price.”
93. Mahmud Ghaznavi was surely acting upon the imperative of transplanting
strong Arabo-Persianate imperial foundations into Zabulistan and contig-
uous regions as he employed the “strong-arm methods” of virtually coerc-
ing scholars and literati to bejewel his court (cf. Bosworth [1963] 2015: 34,
129–130; Bosworth 1968a: 36, 38 and passim). See also Meisami 1999: 50ff. for
a discussion of the possible reasons behind Mahmud’s ambivalence toward
Persian literature.
94. The areas of nomadism and seasonal transhumance were discussed in Chapter
2. For the Ghuris (and among them the future Shansabanis), their garm-sir
areas in particular – in southern Ghur and Zamindawar, perhaps touching
upon Sistan – would likely have offered a number of shrines or ziyarat dating
from at least the eleventh century and likely earlier, though each of them
surely fluctuated over the centuries in terms of their ability to attract pilgrims,
depending on climate and shifts in river courses, among other unforeseeable
factors. Cf. Ball 2019: e.g., Nos. 597/1114, 1264/2182, 1267.
95. For twentieth-century nomadic interactions with the shrines around
Chisht, see esp. Utas 1980: 64. According to Khazeni (2012: 147–148), in the
eighteenth–nineteenth centuries Central Asian Türkman nomads integrated
Naqshbandi Sufi tombs in their migrations, particularly during their seasonal
contacts with towns or market centers for commercial purposes. See also
supra in this chapter. The reverse process of murshids traveling to see their
disciples in pastoralist communities – usually in the autumn, the traditional
period of socializing when the harvest was in – was documented among the
Helmand Baluch: cf. Amiri 2020: 153.
96. This was the case notwithstanding Green’s (2019: 17) observation that “the
gradually increasing numbers of [Saljuq- and Ghaznavid-patronized] madra-
sas and khanaqahs spread the use of written Persian across new geographical
frontiers” (my emphasis): such proliferation of institutions of learning did not
necessarily create greater Persian or Arabic literacy across lifeway frontiers,
i.e., the nomadic/transhumant populations.
97. The absence of Persian on Shansabani-patronized architecture was noted
by O’Kane (2009: 30) and Allegranzi (2019b [forthcoming], 2020), both of
whom compared the Shansabani inscriptional corpus with that of the sophis-
ticated and courtly Ghaznavids. By the time of the Ghaznavids’ ascendancy in
the eleventh century, Persian – both in verse as well as prose – had undergone
a centuries-long rise alongside Arabic, particularly in courts of the eastern
Islamic ecumene. By the dawn of the new millennium, specifically modern or
“new” Persian had made inroads into the monopoly Arabic once had as the
language of religious and literary textual production (cf. also Allegranzi 2015:
26–30, 32–35; Allegranzi 2019a, vol. I: 19–23; Green 2019: 9–10; and supra in
main text).

CHAPTER 5
The “Ports of India”: Ghazna and

Bust-Lashkari Bazar
W e witness in this chapter the coalescence of the third Shansabani

lineage at Ghazna, and the architectural manifestations of this
process at the erstwhile Yamini capital and its contiguous areas. Stylistic
developments evidenced in the architectural fabrics of the principal palace at
Ghazna (Figures 5.1–5.7), and of many of the built remains at Bust (Figures
5.9–5.13) and Lashkari Bazar (Figures 5.14–5.28), serve as plausible evidence
of extensive Shansabani-period renovations there, most probably commis-
sioned by Mu‘izz al- Din himself. There is very little evidence, however, of
new construction. From the surviving remains, we might surmise that, at
least among some segments of the Ghazna Shansabani elites, the large-scale
interventions at all of these sites were indications of a shift toward greater
sedentism – perhaps better described as an emerging tendency toward res-
idence in permanent rather than textile structures. These indices of change
appear to have been carried even further upon the Shansabanis’ entry into
the north Indian duab, and will be treated in detail in Volume II.
Juzjani cursorily noted Mu‘izz al-Din’s summer and winter capitals (dar
al-mulk-i tabistan and dar al-mulk-i zamistan) as Ghazna and Khurasan,
and eventually Lahore and Hind (respectively). Thus, the sultan’s resi-
dence was not limited to a principal city, but rather also encompassed
its vicinity, not unlike the Saljuqs’ royal encampments around Isfahan
and other major urban centers.1 Such an arrangement could have accom-
modated the requirements of both imperial expansion and the echoes
of a transhumant lifestyle, facilitating campaigns in the respective areas
where climate and resources were amenable. It is worth re-emphasizing
that the Yamini–Ghaznavids also changed locations according to summer
and winter: they had tended to avoid the cold and snowy winters of
Ghazna, instead undertaking campaigns to the north Indian plains when
the oppressive heat of summer was absent there.2 It might appear that the
Ghazna Shansabanis’ movements followed in the seasonal footsteps of
their predecessors. But in fact, these two dynastic elites converged upon
these modi operandi through vastly different historical trajectories – the
Shansabanis emerging from nomadism/transhumance within their scions’
lifetimes, and the Ghaznavids adapting an inherited Perso-Islamic courtly
culture to their specific circumstances (cf. Chapter 1).

238 Iran to India
It also bears reiteration that, throughout this book’s analyses of archi-

tectural groups, prior scholarship has been invaluable as a starting point.
Similarly here, the Italian, French, and other surveys and excavations at
Ghazna and Bust-Laskhari Bazar obviate the need for exhaustive descrip-
tions; they provide the solid bases for tracing the transformations of
Ghazna and Bust-Lashkari Bazar into “new” imperial spaces.
Eastern Impulses
The long “house arrest” inflicted on the sons of Baha’ al-Din by the
infamous ‘Ala’ al-Din Husain Jahan-suz, their paternal uncle, ended only
after the latter’s death: perhaps helping to heal the familial breach, ‘Ala’
al-Din’s son and successor Saif al-Din freed his cousins, even inviting
the elder Shams al-Din into his retinue, while Shihab al-Din went to
yet another paternal uncle, Fakhr al-Din Mas‘ud, the patriarch of the
Bamiyan lineage.3 Fatefully, Shams al-Din participated in the campaign
against the Oghüz in Gharjistan in ah 558/1163 ce, when Saif al-Din
met his tragic death at the hands of the treacherous ‘Abbas ibn Shish. It
was this unforeseen event that led to Shams al-Din’s crowning as Sultan
Ghiyath al-Din at Firuzkuh in the same year (cf. Chapter 1).
Shihab al-Din’s trajectory toward rulership was more protracted and
uncertain. Juzjani implied to his readers that a stinging rebuke by Fakhr
al-Din Mas‘ud of his nephew’s complacency finally stirred the young
man’s ambition. Upon sensing Fakhr al-Din Mas‘ud’s deep displeasure,
Shihab al-Din apparently left Bamiyan for his newly crowned brother’s
court at Firuzkuh, where he surely had hopes of attaining independence
and even distinction. When these hopes were not immediately fulfilled,
Shihab al-Din withdrew to Sistan, taking refuge for about a year in Zaranj
at the court of Malik Shams al-Din Nasri (r. ah 559–564/1164–1169 ce) –
apparently a cruel tyrant to his own people, but eager to preserve good
relations with his overlords the Firuzkuh Shansabanis.4 Sultan Ghiyath
al-Din coaxed his younger brother back to Ghur and bestowed on him
the Istiya area, which lay in the vicinity of Tiginabad/Old Qandahar (cf.
Chapter 3): it had been the appanage of the late patriarch Saif al-Din Suri
(d. ah 543/1149 ce) during the early days of the Shansabanis’ regional dom-
inance. Eventually, Shihab al-Din was given the bigger prize of Tiginabad/
Old Qandahar itself. But this territory came with challenges due to its
proximity to the Oghüz encampments in Ghazna and its environs.
With historical hindsight, it becomes clear that avuncular rebuke alone
was not responsible for Shihab al-Din broadening his horizons, just as
appeasing sibling tensions in itself was insufficient for Sultan Ghiyath
al-Din to mobilize Firuzkuh’s forces toward the conquest of Ghazna. In
addition to these intimate impetuses, a sequence of specific events had to

The “Ports of India”: Ghazna and Bust-Lashkari Bazar 239
converge for the Shansabanis’ long-awaited triumph over the erstwhile

Yamini capital. In what could be termed a historical irony, it was the
Oghüz who provided a pivotal catalyst for this victory: they had not only
occupied Ghazna by c. ah 556/1161–1162 ce, their further encampments
and disruptive movements ranged beyond Zabulistan into Zamindawar
(map, p. xvi).5 Thence, these disruptions touched Shihab al-Din’s new
territories of Tiginabad and its vicinity. His repeated confrontations with
the Oghüz in turn brought him in contact with the once great capital: as
the apparent core of the Oghüz’s regional presence, the city necessarily had
to have their encampments eradicated for a final solution to the “Oghüz
problem.”6 In ah 568–569/1173 ce, the definitive routing of the Oghüz
from Ghazna – the “Port of India” – established a third Shansabani lineage
with the crowning of Shihab al-Din as Sultan Mu‘izz al-Din.7
Past scholarship has tended to absorb the Ghazna Shansabanis into an
overarching “Ghurid” dispensation. But considering them as a deputa-
tion acting on the orders of (or otherwise subordinate to) the Firuzkuh
Shansabanis leads to a historical misperception: based on this assump-
tion, the renown of the successful India campaigns usually accrues to the
Firuzkuh branch as the senior lineage.8 In Chapters 1 and 3, we already
took note of the multiple geo-cultural, socio-economic, and indeed
unforeseen historical factors – along with the possibility of intimate emo-
tional interventions – that collectively presented alternatives to the textual
interpretations (such as Juzjani’s) of Shansabani expansion as a process
centrally orchestrated from Firuzkuh. In addition, throughout the preced-
ing chapters we have also seen the powerful physical indices of centrifugal
tendencies among the Shansabani lineages, as they crafted imperial infra-
structures in keeping with the differing cultural geographies they each
inhabited.
Although a small market town in the mid-first millennium ce, Ghazna
had grown to a sizeable commercial center by the tenth century. The set-
tlement was the principal waystation on the Kabul–Qandahar route, with
access to the Kurraman and Gomal passes (cf. Chapter 1) and the middle
Indus’s alluvia (map, p. xv). Moreover, with the Rutbils’ dominance of
Zabulistan and the contiguous region of Zamindawar possibly as late as
the tenth century, Ghazna also became a major Buddhist monastic center.
As was the case with Bamiyan, the site’s command of an important node
within a political geography combined with its Buddhist aspects, making
it a busy commercial and pilgrimage center “securing the Central Asian
and Iranian relations with India.”9
It would not be unreasonable to propose, then, that the Shansabanis
of Ghazna underwent a further differentiation as an increasingly inde-
pendent lineage, akin to the Shansabanis of Bamiyan. Similar to their
agnates, the Ghazna Shansabanis inhabited a distinct historical–cultural
geography within Afghanistan: rather than the links of Firuzkuh with

240 Iran to India
greater Khurasan and its commercial emporia to the west, or the centrality
of Bamiyan as the nodal convergence of east–west and north–south com-
munication routes, the commercial and cultural networks encompassing
Ghazna and the region of Zabulistan in general had long been oriented
toward the east and India (see Chapters 2 and 3). For Mu‘izz al-Din
and his loyalists and dependents, then, the conquest of Ghazna in itself
was not enough; as had been the case for the Yamini–Ghaznavids before
them, the imperial machinery established here could be sustained only if
the location was used as a springboard for expanding eastward (see also
Chapter 6).
All in all, once again we see that the Shansabani lineages of the 1170s
ce would be better understood as a confederacy rather than a singular unit,
with the three agnatic branches having disparately expanding ambitions
(map, p. xv), but nominally subject to overlordship. Although anthropo-
logical analysis of modern tribal organizations in Afghanistan is plentiful –
perhaps most notably by the “accidental” architectural historians Glatzer
and Casimir – the question of confederation has been directly addressed
among the Basseri tribe and its larger Khamseh confederacy, their periodic
rivals the Qashqai confederacy, and the Bakhtyari tribes – all nomadic
pastoralists in south-central Iran. 10 While these studies have highlighted
differing circumstances and equally varied processes of confederation, they
also underscore confederation as inherent to the nomad–urban contin-
uum: kinship-based and other groups tended to come together, that is,
confederate, as a response to shifting circumstances, often at the instiga-
tion of a leader and his lineage.11 The Bakhtyari example may provide the
most useful parallel to the Shansabanis, as even in regions where imperial
presence was not constant or strongly perceptible – certainly the case in
Ghur – “confederations form[ed] … in response to an external stimulus –
typically, a need for common defense or an opportunity for expansion or
conquest”.12
There are some indications that Mu‘izz al-Din only grudgingly acknowl-
edged the seniority of the Firuzkuh branch of Shansabanis.13 Moreover,
Ghiyath al-Din’s ambitions of expansion into Khurasan apparently pre-
cluded any independent share therein for the Shansabanis at Bamiyan,
or for the lately established Ghazna lineage embodied in Mu‘izz al-Din,
his dependents and military followers. The Ghazna Shansabanis had little
choice but to direct their energies toward India. Mu‘izz al-Din and his
ghulams were now oriented eastward, at least in part swept up in the
geo-historical momentum of their predecessors, the Yamini–Ghaznavids.
Collectively, Mu‘izz al-Din’s military successes – and arguably, also his
reverses – were to have more palpable, local, and transregional ramifica-
tions over the ensuing millennium.

Ghazna and Bust–Lashkari Bazar14
The multi-faceted significances of the Shansabani conquest of Ghazna

cannot be sufficiently emphasized. This event was pivotal in the
Shansabanis’ overarching historical trajectory: the taking of the city cer-
tainly resulted in the irretrievable (if painfully gradual) demise of the
Yamini–Ghaznavids, and the eradication of the Oghüz nomads from
Zabulistan and Zamindawar introduced new horizons for Shansabani
territorial expansion. And as already noted, the establishment of the col-
lateral Shansabani lineage at the old Yamini capital not only expanded the
Shansabani confederacy, it opened the way to India.
This chapter argues that the conquest of Ghazna could have had addi-
tional, meaningful consequences on a more intimate plane: the architec-
tural remains at Ghazna and its attendant regions evince signs of a shift in
lifeways for the new Shansabani lineage vis-à-vis their Firuzkuh cousins – a
shift toward greater sedentism, which would also echo in the architectural
remnants of their campaigns beyond the Indus into the north Indian
plains (see this Volume’s Introduction, and Volume II).
However, akin to the archaeological remains at Jam-Firuzkuh (cf.
Chapters 3 and 4), any analysis of the documented architectural fabric as
well as the still to be studied material remains of Ghazna and Bust-Lashkari
Bazar also requires important caveats. It bears reiterating that the Mongol
campaigns along the Zabulistan–Zamindawar corridor (map, p. xvi)
and their various aftermaths, unfolding during the third through eighth
decades of the thirteenth century, certainly disturbed what had survived of
these urban and palatial sites’ historical remnants.15 In addition, given their
multiple occupational layers, the Ghaznavid sites of southern Afghanistan
comprised vast archaeological complexes requiring collaboration among
a range of specialists from the pre-Islamic centuries onward. The sheer
scale of these sites has meant that, despite many seasons of survey and
excavation – cumulatively spanning the 1920s through 1970s16 – they
were worked upon only partially before political and military insecurity in
Afghanistan made work there extremely difficult, if not impossible. Thus,
like Jam-Firuzkuh, what is presented here must be viewed as provisional,
until exploration in these areas can be reinitiated.
There are nonetheless important observations that can be made.
Archaeological and art-historical analyses of the remains of Ghazna and
Bust–Lashkari Bazar certainly revealed distinguishable phases of architec-
tural intervention at these localities. As discussed in detail below, the gen-
erally accepted sequence of events is that the existing Ghaznavid-period
buildings and complexes at Ghazna and Lashkari Bazar – Bust having been
less studied – suffered partial destruction by fire. At Lashkari Bazar the
destruction was evidenced on two occasions, the first attributable to ‘Ala’
al-Din Jahan-suz’s sacking in ah 544/1150 ce, and the second perhaps to

242 Iran to India
the Mongol incursions of the early to mid-thirteenth century, after which

apparently neither of these sites fully revived.17
The series of events recorded in Juzjani’s textual narrative seem to
concord with the evidence presented by material culture – further dis-
cussed below. The author’s account of the destructive visitations of ‘Ala’
al-Din Husain Jahan-suz upon Ghazna and Lashkari Bazar in ah 544/1150
ce, could have resulted in the first signs of destruction by fire at both
sites.18 The partial date of ah 55x/1156–1165 ce in Lashkari Bazar’s South
Palace inscription (Figure 5.17) (cf. below) would then indicate interven-
tion(s) at least at this site after the destruction. As outlined in Chapter
3, in ah 553/1158 ce ‘Ala’ al-Din Husain had demanded and eventually
wrested Tiginabad/Old Qandahar from the recently defeated Ghaznavid
sultan Bahram-shah. This Shansabani territorial gain effectively disrupted
the Ghazna–Zamindawar route, preventing the later Ghaznavid sultans
from easily traveling to Bust and Lashkari Bazar. The fragmentary date
of Lashkari Bazar’s South Palace inscription, then, could refer to work
undertaken sometime after ah 553/1158 ce, when ‘Ala’ al-Din prevailed
yet again over the Ghaznavid forces; this date falls within the decade-long
span of the incomplete inscription.19 But, as argued below, the substan-
tial and large-scale renovations at Lashkari Bazar should be re-dated to a
period after the final Shansabani takeover of Ghaznavid territories as of
c. ah 569/1173 ce and onward.
Ghazna
Recent analyses focused on the marble, terracotta, and stucco architectural
fragments ensuing from the Italian excavations at Ghazna – conducted
more than a half-century ago (1950s–1960s) – are finally creating a fuller
picture of Shansabani interventions here. Prior to these new studies,
Shansabani traces at Ghazna appeared to have been extremely meager,
limited to a reoccupation of an elite residence in the Rauza hills, overlook-
ing the city’s central area (Figure 5.1): here, among the many rediscovered
objects was a store of remarkably well-preserved lusterware, which ulti-
mately gave their name to the site as the House of Lusterware. The objects
were collectively interpreted as dating to the late twelfth or early thirteenth
century, definitely after the first Shansabani sack of Ghazna in ah 544/1150
ce, and likely also after the final victory of ah 569/1173 ce.20 However,
ongoing studies of terracotta architectural decoration, recovered from
Ghazna’s central palace, profoundly alter the perception of an ephemeral
Shansabani presence in the city. It seems that the Shansabani interventions
at the palace were more commensurate with their momentous occupation
of the capital, and especially the crowning of Shihab al-Din as Sultan
Mu‘izz al-Din, who effectively appropriated – though with some needed
territorial consolidation – the central Ghaznavid holdings encompassing

1
2
A
B
Central Palace
A. Rectangular Enclosure south of Palace
B. Trapeziform Enclosure
1. Minaret of Bahram Shah
2. Minaret of Masud III
0 50 100 200 300 400 500 600 700 meters
Figure 5.1 Ghazna, plan from aerial photograph of western part of city.
From Scerrato 1959: fig. 17.
Zabulistan and Zamindawar, through the easternmost borders of Sistan

(maps, pp. xv and xvi).
The central palace of Ghazna (Figures 5.1 and 5.2), to the east of the
minar of Mas‘ud III – and surely falling in the shadow of the originally
much taller tower (Figure I.20) – was previously attributed to this sultan
through a web of tangential associations. But recent re-examination of
the archaeological data from this vast complex has led to a new under-
standing of the palace, encompassing a considerably more complicated
architectural development likely beginning at least a half-century before
the reign of Mas‘ud III (r. ah 492–509/1099–1115 ce), and unfolding via
several identifiable phases.21 Ongoing studies of the palace’s terracotta and
stucco architectural decoration strongly suggest that a visible and perhaps
substantial mark was left there in the wake of the Shansabani victory of
ah 569/1173 ce as well.
More than three thousand terracotta and nearly seven hundred stucco
elements from the palace’s decoration – many with epigraphy – are under-
going detailed study and recontextualization within the palace’s various
building phases (Figures 5.3–5.5). These decorative components still carry
traces of paint, indicating their original polychromy.22 Among the ter-
racotta remains, several broken pieces retrieved from the northeastern
and southern zones of the palace’s large central courtyard carried deci-
pherable epigraphic fragments. Some inscriptions were simple doxologies
(e.g., al-mulk li-llah), echoing the simplicity of the inscriptional frieze
likely running along the walls of the palace’s principal northern entrance.23
However, other fragments have been read as being part of a longer frieze
containing the kunya and laqabs of Sultan Mu‘izz al-Din.24
These fragments bear an undeniable resemblance to the partial epigraphic

244 Iran to India
Figure 5.2 Ghazna, central palace, plan of excavated areas and hypothetical
reconstructions. After Giunta 2020: fig. 2.
bands still in situ during the Italian excavations of the 1960s (Figure 5.6):
the chunky, bold letters with decorative stem terminals standing out against
a background of small, triangular solids and voids – emphasized further by
being painted. Moreover, these terracotta epigraphic bands formed a stark,
formal contrast to the refined marble dado below (Figures 5.6 and 5.7),
which had consisted of rectangular panels carrying arabesques and floral
motifs bordered by a continuous horizontal band of elegant floriated Kufic
or naskh25 (cf. below). Thus far, then, analyses of the painted terracotta
epigraphic fragments from Ghazna’s central palace point to very marked
renovations of at least parts of the main courtyard’s façade, probably exe-
cuted during the reign of Mu‘izz al-Din (r. at Ghazna ah 568–602/1173–
1206 ce) and perhaps even shortly after the Shansabani conquest of the city.
Indeed, Mu‘izz al-Din’s notable renovations of the Ghazna palace

Figure 5.3 Ghazna, central palace, fragment of painted stucco decoration.

From Allegranzi 2021: fig. 3, detail 1 (repr. of IsMEO inv. no. C5612).
Figure 5.4 Ghazna, central palace, fragment of painted stucco decoration.

From Allegranzi 2021: fig. 3, detail 2 (repr. of IsMEO inv. no. C2719).

246 Iran to India
Figure 5.5 Ghazna, central palace, fragment of painted stucco decoration. From
Allegranzi 2021: fig. 3, detail 3 (repr. of IsMEO inv. no. C5784).Ghazna, central
palace, fragment of painted stucco decoration.
Figure 5.6 Ghazni, palace of Mas‘ud III (1099–1115), excavation site, dado
remnants. © Photograph Josephine Powell c. 1960. Harvard Fine Arts Library,

Figure 5.7 Detail of Figure 5.6, Ghazni, palace of Mas‘ud III, juxtaposition of
Ghaznavid- and Shansabani-period revetments.
were not limited to the additions of large, bold terracotta epigraphy tow-
ering above their predecessors’ more refined marble dadoes. The 1950s
Italian excavations at the site also unearthed approximately 160 glazed
tiles, square or polygonal in shape, with molded decorative motifs. The
glazes on the tiles were varied, their colors including green, yellow, brown,
red, blue, and turquoise of various hues. The molded decoration encom-
passed a wide array of geometric, floral-arabesque, zoomorphic, and
pseudo-epigraphic motifs. The relatively small number of these glazed tiles
led the archaeologists to suggest they were used as punctual decoration,
inspired by grid patterns on woven textiles of the time, which framed
various floral, geometric, and animal motifs. Given the tiles’ retrieval from
the uppermost layers of the Ghazna palace, as well as a few at the House of
Lusterware, Scerrato26 opined that the tiles were manufactured most prob-
ably in the late twelfth and early thirteenth centuries, thanks to a local tile
industry that had apparently survived ‘Ala’ al-Din’s onslaught, the Oghüz
occupation, until the definitive Shansabani conquest of the city. Thus, the
new Shansabani sultan of Ghazna, seeing personified before him the value
of monumental and distinctive imperial patronage, was reviving the glory
of his newly acquired “capital” in his own way (cf. below and Chapter 6).
Bust
The expansive stretch southwestward from Ghazna to the citadel and
settlement of Bust had long marked a busy thoroughfare. At least from
Parthian times or the early centuries of the Common Era, the town of
Bust and its imposing citadel had functioned as an important waystation,
from whence branched a northward route into Zamindawar, touching

248 Iran to India
upon southern Ghur; and another one eastward toward Tiginabad/Old

Qandahar, and ultimately back to Ghazna again. Possibly since the time of
the Samanids (tenth century), and almost certainly since that of Mahmud
Ghaznavi’s father Sabuktigin (d. ah 389/999 ce), a series of large residen-
tial complexes – and likely a few commercial buildings – had appeared on
the Helmand’s left bank several kilometers north of Bust, essentially creat-
ing an identifiable area ex urbis. This exurb’s historical name of al-‘Askar –
now Lashkari Bazar, discussed below – could indicate that, during various
moments prior to the eleventh century, the vast plain had been used for
military encampments, which were dependent on nearby Bust.27
Given the Shansabani presence at Tiginabad/Old Qandahar as of the
later 1150s (cf. above and Chapters 1 and 3), control of the entire corri-
dor from Ghazna through Bust would have been advantageous, and was
probably accomplished sometime shortly after the Shansabanis’ fateful
victory at Ghazna in ah 569/1173 ce.28 Ensuring this region’s integration
firmly within the Shansabani ambit was not only impelled by its previous
Ghaznavid connections; securing the Zabulistan–Zamindawar corridor
was surely necessary for the newly titled Sultan Mu‘izz al-Din’s further
kingly aspirations and territorial ambitions.
Bust itself was a significant fortified town, following the regionally
widespread layout of a royal enclave (Pers. arg) ensconced within a larger
walled settlement (Pers. shahristan) (Figure 5.8). Within these type of plans,
the more secure enclaves were frequently at the outer edge of the citadel,
Figure 5.8 Bust, Helmand Province, Afghanistan, general plan. From Ball 2019:
No. 149.

Figure 5.9 Bust, citadel and arch, general view from northeast. © Photograph
and perhaps even naturally protected on one flank by a river or cliff.29 In

the case of Bust, the layout followed the rising topography: two concentric
enclaves were placed at the walled settlement’s northwest corner, with the
innermost enclave at a high promontory overlooking the town and plains
below (Figures 5.8 and 5.9).30 The enclosed enclaves decreased in accessibil-
ity, with the royal palace and other imperial machinery probably located
within the highest and best protected area (Figure 5.9).
Understandably, both Bust and Lashkari Bazar had received lavish archi-
tectural attention from at least two of the great Ghaznavid sultans, Mahmud
(r. ah 389–421/999–1030 ce) and his son Mas‘ud I (r. ah 421–432/1030–1041
ce) during that dynasty’s period of apogee. As these areas continued to play
a significant role in the governance and control of Zamindawar and con-
tiguous areas and byways, Mu‘izz al-Din would have likely embarked on
significant interventions there after the conquest of Ghazna in ah 569/1173
ce. But the nature of these interventions can only be tentatively understood
with the currently available archaeological and other data. Further, based
on the Shansabani patronage at Ghazna discussed previously – and on what
we will see at Lashkari Bazar below – it is reasonable to propose that at Bust
also, there was extensive reuse of the pre-existing Ghaznavid-era structures.
New constructions were less frequent, however, as detailed forthwith.
The location of the mysterious “galleried” well within the upper enclave
of the Bust citadel underscores this structure’s importance (Figures 5.8,
5.10 and 5.11). Several of the well’s seven stories were underground and
led to a water catchment area.31 However, no epigraphic finds or other
chronological indicators have been associated with the structure, so that
we can only conjecture as to its foundation date and period of use. While

250 Iran to India
Figure 5.10 Bust, citadel, ruins, interior, seven-story well: man stands among upper
floor arches. © Photograph Josephine Powell c. 1960. Harvard Fine Arts Library,
predictably constructed of mud and baked brick, the well’s extremely

rare if not unique configuration appears to have no precedents in western
Eurasia, but its form and function are comparable to the numerous stone
stepwells throughout the Indian subcontinent.
The stepwell tradition extended from the seventh through at least the
sixteenth centuries throughout the subcontinental region – and evidently
also included Muslim patrons of what were essentially “civic” works for
community benefit.32 Numerous stepwells dating to the eleventh–twelfth
centuries still survive throughout the north Indian duab, and would have
surely been encountered during both the Ghaznavids’ (c.1000–1040 ce)
campaigns into northern India, as well as the Shansabanis’ own forays as of
the late 1170s, and more consistently in the 1190s – either or both of which
could have served to transport the general concept from Hindustan to
Zamindawar. Given the Bust structure’s overall proportions and structural
affinity with Ghaznavid- and Shansabani-patronized architecture, the well
could be tentatively dated to the eleventh–twelfth centuries.33

Figure 5.11 Bust, citadel, ruins, interior, seven-story well: man stands among upper
floor arches. © Photograph Josephine Powell c. 1960. Harvard Fine Arts Library,
The oft-illustrated, distinctive monumental arch inside the lower

enclave of Bust (Figures 5.9, 5.12 and 5.13] would appear to be one of the
few, new Shansabani foundations at Bust, perhaps along with the galleried
well just discussed. The arch would certainly qualify as a project of impe-
rial proportions, though its context and ultimately the impetus behind
it both remain mysterious.34 The arch’s interpretation as a monumental
entrance addorsed to a large mosque or other complex was refuted early
on by the fact that elaborate stucco decoration graced its eastern and
western surfaces, indicating its viewability from both sides (Figures 5.12
and 5.13).35 Its more elaborately decorated eastern face carried Q. II: 127,36
and, depending on the calligraphic spacing, possibly other verses as well.
This part of Surat al-Baqara describes Abraham
… raising the foundations of the House and Ismael [saying], “Our Lord,
accept this from us. Indeed You are the Hearing, the Knowing.”

252 Iran to India
Figure 5.12 Bust, arch, south pillar, carved decoration. © Photograph Josephine
Powell c. 1960. Harvard Fine Arts Library, Special Collections.

Figure 5.13 Bust, arch, front face and intrados, north, carved geometric decoration
and inscription, detail. © Photograph Josephine Powell c. 1960. Harvard Fine
On stylistic analysis particularly of the arch’s epigraphy and geometric

decoration, Sourdel-Thomine dated it to the third quarter of the twelfth
century, basing the date range on comparisons with the epigraphy of
Shah-i Mashhad.37 As we saw in Chapter 4, the period of construction
for that madrasa complex likely spanned a decade, from ah 561 to 571
(1165–1176 ce). Thus, a date for the arch at the end of this range (c. ah
570/1175 ce) would not be implausible.
Several aspects of the Bust arch suggest that it served as a triumphal
monument commemorating an event falling within Shansabani ascend-
ancy in the region. In his re-surveying of the Bust citadel, T. Allen
proposed that the arch was part of a reconfiguration of the principal, cere-
monial entrance into the lower enclave, which in itself served as a gateway
to the upper more insulated area.38 The arch was thus a public – even
“civic” – structure, and also evinced an association with royalty and its
ceremonial expression. With the proposed date of c. ah 570/1175 ce for the
arch as a basis, it is conceivable that this monument was erected in virtual

254 Iran to India
simultaneity with the Jam-Firuzkuh minar, acting as the other half of the
dyadic declaration of the Shansabanis’ final triumph over the Ghaznavids
and the Oghüz, and the possession of the former’s imperial centers (see
Chapter 4). But rather than memorializing a Shansabani triumph that
essentially acknowledged Ghiyath al-Din’s seniority as at Firuzkuh, the
Bust arch appeared to focus on the newly crowned Sultan Mu‘izz al-Din.
Given the far-reaching ramifications of this fateful victory, it is more than
likely that some type of memorialization of it was also erected at their
enemies’ old capital of Ghazna but has since been lost.
Furthermore, as was the case with the Jam-Firuzkuh minar – discussed
in Chapters 3 and 4 – the Bust arch’s surviving epigraphy also stands out
for the uncommonness of its content: according to Blair’s authoritative
study of the monumental inscriptions of the eastern Islamic lands, Q.
II: 127 and its proximate verses were not documented in these regions,
at least through the early twelfth century, and they appeared but rarely
elsewhere.39 But the ‘ulama’ and other intellectuals within Mu‘izz al-Din’s
ambit could easily find Qur’anic verses appropriate to any occasion. Given
the rarity of these particular verses, it seems that they were selected with
specific intent. The actual message of the verse(s) is an invocation of divine
blessing on a new “house,” even a royal house – especially apropos here.
The monument overall, then, could conceivably be proclaiming the estab-
lishment of the third Shansabani lineage and the new rule of Sultan Mu‘izz
al-Din throughout the Ghaznavids’ erstwhile territories.
Lashkari Bazar
Very few firmly dated architectural elements – which were not portable,
unlike coins, ceramics, and other objects – have been recovered at any of
the three sites forming the subject of this chapter. One significant find
was the inscriptional panel affixed to the southern façade of Lashkari
Bazar’s great South Palace (Château du Sud), in situ during the French-led
excavations there in the 1960s (Figures 5.14, 5.15, 5.17): it bore the partial
date of ah 55x, corresponding to the decade of 1156–1165 ce. Based on this
dating, the renovations at Lashkari Bazar have been attributed primarily
to the decade after 1150 ce, an attribution I propose to modify in the rest
of this chapter.
Sourdel-Thomine’s rigorous analyses of the architectural epigraphy and
decoration at Bust and Lashkari Bazar resulted in compelling – if at times
debated – conclusions.40 Her deep knowledge of epigraphic styles and dec-
orative iconographies, spanning virtually the entire Persianate world of the
eleventh through fourteenth centuries, permitted convincing comparisons
between the inscriptional and ornamental programs of Lashkari Bazar’s South
Palace and relevant Saljuq, Ghaznavid, and Shansabani parallels. Also taking
into account the South Palace inscription’s partial date of ah 55x/1156–1165 ce

(Figure 5.17), she attributed the

renovations at the Palace to
the years following ‘Ala’ al-Din
Husain Jahan-suz’s destruc-
tive path through the region of
Ghazna and Zamindawar.41
Sourdel-Thomine systemat-
ically identified the following
mid-twelfth-century changes
there (Figures 5.14 and 5.15):
additions of epigraphic and
other decorative elements to
the Palace’s principal southern
façade, which Schlumberger
contrasted to the “nudity” of
the other façades (Figure 5.16)42;
a mihrab in the Mosque of the
Forecourt (Figure 5.18), part of
the long esplanade approaching
the Palace from the south; ren-
ovations of all four façades of
the Palace’s large central court
(Figures 5.14, 5.19); the stucco
revetment, epigraphic program,
and mihrab of the oratory of Figure 5.14 Lashkari Bazar, Helmand Province, South
Hall I (Figures 5.20–5.22), north Palace, plan of excavations. From Schlumberger 1978: pl. 4.
of the central court; the stucco
imitating cut brickwork in Apartment II, northeast of the central court;
and the molded stucco and incised geometric decoration in Building XX,
a northeastern annex to the Palace. Sourdel-Thomine also attributed to
Shansabani patronage a new decorative program in Residence XIII (Figure
5.23), the so-called “Palace of Racquets,” close to Bust on its northeastern
flank. Since very little other definitively datable evidence has been recov-
ered elsewhere in this vicinity, ceramic and epigraphic finds – together
with archaeological stratigraphy and references in historical texts – have
been the principal clues for the aforementioned chronological sequence.43
With much less information at her disposal than we have now,
Sourdel-Thomine applied the South Palace inscription’s partial date of ah
55x/1156–1165 ce (Figure 5.17) to all later interventions within the complex.
According to this chronology, the works would have taken place within
the period immediately after ‘Ala’ al-Din Husain Jahan-suz’s mpaign of
destruction at the site and throughout the Zabulistan–Zamindawar corri-
dor in ah 545/1150 ce. To be sure, the partially surviving inscription could
have referred to some work undertaken at the complex during this decade.

256 Iran to India
Figure 5.15 Lashkari Bazar, South Palace, southern façade. From Sourdel-Thomine
1978: pl. 52b.
Figure 5.16 Lashkari Bazar, Summer (or Grand) Palace, ruins, exterior, southern
wall, detail. © Photograph Josephine Powell c. 1960. Harvard Fine Arts Library,

Figure 5.17 Lashkari Bazar, South Palace, south portal detail.

From Sourdel-Thomine 1978: pl. 54d.
But given the long list of renovations, and particularly the nature of some
of them, it is improbable that all the projects listed above were executed
in the late 1150s. Major renovations would have been unlikely just a few
years after ‘Ala’ al-Din Jahan-suz’s willful destructiveness there and in its
environs, even though he might have ultimately planned to occupy the
area – along with Ghazna itself – but ultimately left this to his next cam-
paign in the region (cf. Chapters 1 and 3).
Angular Kufic doxologies, such as al-mulk li-llah, were added to the
palace’s main southern façade (Figures 5.15 and 5.16), as well as the interior
court façades (Figures 5.14, 5.19). Such simple epigraphy was a far cry
from the sophisticated verses on the dadoes of the Ghazna palace’s central
court, composed in various meters of the Persian poetic tradition and
possibly glorifying the Ghaznavids’ dynastic lineage and expounding legal
and theological tenets. But at least in content – the rarity of in situ remains
at Lashkari Bazar prevents stylistic comparison – the simple sayings were

258 Iran to India
Figure 5.18 Lashkari Bazar, South Palace, Mosque of the Forecourt

(photograph c. 1950s). © Courtesy of Musée des Arts Asiatiques-Guimet, Paris.
Figure 5.19 Lashkari Bazar, South (or Grand) Palace, ruins, courtyard.
© Photograph Josephine Powell c. 1960. Harvard Fine Arts Library,

Figure 5.20 Laskhari Bazar, South Palace, Oratory F, north of central courtyard.
From Schlumberger 1952: pl. XXXII, 3.
in keeping with the later renovations of the same Ghazna palace (Figures
5.6 and 5.7), credited to Mu‘izz al-Din (see above).44 It would not be
unreasonable to suggest, then, that the addition of these new decorative
programs on the various surfaces of the palace were undertaken within
the last quarter of the twelfth century (c. ah 570/1175 ce), after the defin-
itive Shansabani conquest of the region, while Mu‘izz al-Din was casting
himself as successor to the Ghaznavids.
A half-century of new studies has added further nuance to
Sourdel-Thomine’s conclusions. Specifically, a distinctive South Palace
renovation can now be more definitively pinpointed to the last quarter of
the twelfth century. The ivan-like yet quite intimate prayer space directly
abutting the magnificent Audience Hall I to its southwest (Figures 5.14,
5.20), known as Oratory F, surely boasted an exquisite decorative program
in its day.45 On its north and south walls were stucco dadoes with inter-
twining geometric lozenges framing arabesque-floral motifs, horizontally
topped by a running frieze beginning with Bismillah on the north wall.
The frieze continued with Q. XLVIII:1–6 (al-Fath) extending onto the

260 Iran to India
Figure 5.21 Lashkari Bazar, South Palace, Oratory F, stucco decoration

(photograph c. 1950s). © Courtesy of Musée des Arts Asiatiques-Guimet, Paris.
south wall, with possibly more verses from the same Sura (Figure 5.21).
Oratory F’s focal point was of course the west-oriented mihrab (Figure
5.22), flanked by several vertical bands of undulating floral arabesques
and epigraphy: the band rectangularly framing the mihrab contained Q.
II:256 (al-Baqara); in the niche’s hood was Q. CXII:1–4 (al-Ikhlas) in
repetition; and simple doxologies such as al-mulk li-llah were repeated
in the niche’s lower part. The overall decorative program of the Oratory
can be aptly characterized as one evincing a horror vacui, “the compulsion
to cover – almost to swamp – every available space of wall surface with
decoration.”46
We saw in Chapter 4 that several Qur’anic verses in Shansabani-
patronized architectural decoration were otherwise rare or undocumented,
both in the eastern Persianate regions and in the wider Islamicate lands.
To be sure, Q. II:256 – the verse following the practically ubiquitous and
revered Ayat al-Kursi (Q. II:255) – also appeared with some frequency
in architectural epigraphy, extending from Palermo to Delhi and span-
ning the tenth through sixteenth centuries ce.47 However, verses from Q.
XLVIII had evidently first appeared in the eastern Islamicate world on the
minar of Mas‘ud III at Ghazna (c. ah 503/1110 ce), and within the mosque
at Barsiyan in the Isfahan oasis, dated by inscription to ah 528/1134 ce.
I subsequently proposed that, in the 1160s–1170s, the Ghazna decora-
tive program had been the underlying inspiration for Shah-i Mashhad’s
epigraphy, particularly the inscriptional program gracing the complex’s
imposing south façade. Intriguingly, we shall see in Chapter 6 that the
Sura’s beginning verses appeared on yet another Shansabani-associated

Figure 5.22 Lashkari Bazar, South Palace, Oratory F, mihrab.

From Sourdel-Thomine 1978: pl. 147b.
building: the Tomb of Sadan Shahid (late twelfth century) near the city of
Multan in the middle-Indus region.48
Equally noteworthy are all four verses of Q. CXII, which were common
primarily in epitaphs from around the western Indian Ocean; however,
as discussed in Chapter 4, the Sura tended to be absent in the eastern
Islamicate world or the Persianate lands through at least the early twelfth
century. It was as of the 1160s that the verses appeared at Chisht, inside
the western domed structure associated with the Firuzkuh Shansabanis –
perhaps appropriately, as the complex likely had funerary associations.
Finally, the abundant use of doxologies, which helped to create the dec-
orative horror vacui of Shansabani-patronized structures, had been well
attested from the mid-twelfth century onward.
Thus, it would appear that Q. XLVIII’s first several verses, along with
the four verses of Q. CXII (often repeated) had become familiar to the
Firuzkuh Shansabanis, and were also employed thereafter in the Lashkari
Bazar renovations, which were within the purview of Sultan Mu‘izz al-Din

262 Iran to India
and the Shansabani lineage newly established at Ghazna. In fact, both

sets of verses came to be integral parts of the architectural decorations of
buildings along the Indus’s banks during the later 1170s–1180s, buildings
constructed in the wake of the eastward campaigns of Mu‘izz al-Din and
the Ghazna Shansabanis.49
The use of these verses within the epigraphic program of the South
Palace’s Oratory F is significant for at least two reasons: since they came
into limited circulation in the eastern Persianate regions – especially
Afghanistan – with the architectural patronage of the Firuzkuh Shansabanis
as of the 1160s ce, their presence at Lashkari Bazar renders unlikely the
earlier date that Sourdel-Thomine suggested, viz. following ‘Ala’ al-Din
Husain’s violent passing through the region more than twenty years
before. Instead, the verses’ appearance points to a date as of the late 1170s
certainly for this part of the South Palace, which would coincide with
Mu‘izz al-Din’s interventions there. But these inscriptions hold additional
importance for our purposes: adopting what could have been a group
of Qur’anic verses coalescing in the architectural patronage of his older
brother, it appears that the newly crowned Shansabani Sultan Mu‘izz
al-Din also left an intentional architectural mark upon his own hard-won
imperial prizes. Indeed, this particular epigraphic association continued
into Shansabani-associated architecture throughout the Indus’s alluvia, as
we shall see in Chapter 6.
A New Imperialism
This chapter’s examinations of architectural remains attributable to the

Shansabanis at Ghazna, Bust, and Lashkari Bazar enable several observa-
tions regarding the emergence of a third Shansabani lineage. Perhaps most
tantalizingly, they reveal the new imperial modus operandi of the recently
crowned Sultan Mu‘izz al-Din and the Ghazna Shansabanis. These obser-
vations will be extremely important as we examine their expansionist activ-
ities farther eastward, initially throughout the plains and valleys intimately
associated with the Indus River (Chapter 6), and ultimately in the north
Indian duab (cf. Volume II).
Upon “zooming out” to a bird’s-eye view of Sultan Mu‘izz al-Din’s
establishment of Shansabani rulership over the Yamini–Ghaznavids’ core
imperial regions, we cannot help but note that, architecturally speaking,
Shansabani authority was expressed primarily via a marked renovation
and reuse of the existing built environment rather than the creation of
it ex novo: palatial and other complexes were not razed and then erected
anew; instead, distinguishable interventions in already present archi-
tectural fabrics declared the latter’s new Shansabani affiliations. Thus,
the commanding courtyard of Ghazna’s principal palace received new

revetments above (Figures 5.3–5.7) – rather than replacing or atop – its

Ghaznavid-period marble dadoes, creating a discernible contrast between
what was already there and what was added in the course of Mu‘izz al-Din’s
renovations. At Lashkari Bazar, the mosque of the South Palace’s forecourt
quite possibly received a new mihrab (Figure 5.18), in this case covering its
Ghaznavid-era predecessor, but the mosque itself was not entirely built
anew. Additionally, this palace’s southern exterior façade (Figures 5.15 and
5.16) and enormous central courtyard (Figures 5.14, 5.19) also acquired new
decorations and revetments, and the intimate oratory abutting Audience
Hall I (Figures 5.20–5.22) displayed Shansabani renewal in its decoration,
perhaps especially in its epigraphic program (cf. above and below). It
was only at Bust that we detected new construction datable to the late
twelfth-century conquest of the region by Mu‘izz al-Din (Figures 5.8–
5.11), but this also occurred amidst what appeared to be pre-existing built
remains that continued in use (e.g., Figures 5.12 and 5.13). Thus, Ghazna,
Bust, and Lashkari Bazar all entered Shansabani possession largely through
appropriation, rather than erasure and new construction.
We might term this specific mode of building empire re-inscription – a
process that we will encounter in the Ghazna Shansabanis’ forays ever
farther eastward (see Chapter 6). Re-inscription encompassed the reuse
and at times repurposing of societal assets such as the built environment,
usually through new endowments (Ar.-Pers. waqf) toward architectural
and infrastructural maintenance made by the newcomers in perpetuity.
Another fundamental component of re-inscription would have been the
imperial prerogative of minting coinage in the name of the victorious
ruler. While continuity with the standards and iconographic fields of
existing coinage preserved stability, the stamping of the incoming ruler’s
titles and name declared the ascendancy of a new imperial authority. ‘Ala’
al-Din Husain Jahan-suz had done just this upon his second conquest of
Ghazna, continuing the typology of the late Ghaznavid ruler Khusrau
Shah’s (r. ah 552–555?/1157–1160? ce) coins but substituting his own titles
and name. His nephew Mu‘izz al-Din followed suit: he maintained the
standards and fields of late Ghaznavid coinage, replacing the titles and
names with his own – perhaps even airing the family’s “dirty laundry”
by revealing to the world his resentment against submission to his older
brother, Sultan Ghiyath al-Din of Firuzkuh.50
Collectively, the architectural interventions at Ghazna and other
imperial Ghaznavid sites indicate not only that enough had survived
in recuperable condition despite ‘Ala’ al-Din Husain’s depredations a
quarter-century before, as well as the decade-long nomadic encampments
of the Oghüz; they underscore a subtle process occurring in plain sight:
re-inscription, rather than erasure and re-building, appears to have been
Mu‘izz al-Din’s preferred method of claiming his conquests and building
empire. To be sure, reoccupation of perfectly usable spaces was economical

264 Iran to India
and expeditious, but it is still significant how little the Shansabanis actually
built anew in the ex-Ghaznavid domains. I argue in Chapter 6 that, as
the Ghazna Shansabanis successfully campaigned eastward, they imple-
mented this preferred mode of building empire through re-inscription
on an exponentially larger scale. Along the valleys and plains intimately
linked with the Indus River, Mu‘izz al-Din’s followers commissioned
small, usually funerary structures, but their conquests of the region at large
were substantiated principally through the re-inscription of entire urban
environments and landscapes, and the minting of coins – all of which had
previously been inscribed with a Yamini–Ghaznavid presence for nearly
two centuries.
Whether through annexation of previously unclaimed (or at least
unmarked) territories, or the above-described process of re-inscription,
a fundamental resource for the implementation of the Shansabanis’
imperial expansion was the repository of religious, theological, and legal
knowledge possessed by the ‘ulama’ and other scholars in their retinues.
Arguably, among the more palpable indices of these intellectual elites’
participation in empire-building were the epigraphic programs gracing
Shansabani-patronized architectural projects. Chapter 4, proposed that,
in addition to the skilled manual labor required in monumental building,
equally skilled intellectual labor was instrumental in crafting the overall
imperial affiliations and specific religio-theological messages these monu-
ments proclaimed.
The architectural complexes associated with the Firuzkuh Shansabanis
exhibited these intellectual laborers’ contributions in the meaningful
religious/theological content, which could have been selected only by
scholars knowledgeable in the Qur’an, Hadith compilations, and exegetical
literature. Thereafter, Mu‘izz al-Din apparently carried forward the inscrip-
tional corpus – albeit small, as it can only be gleaned from the Firuzkuh
Shansabanis’ building projects that still survive – into Zabulistan and
Zamindawar as he laid claim to the Yamini–Ghaznavids’ imperial centers.
Moreover, the nusualness of the epigraphic content on the magnificent
arch at Bust could indicate the intellectual elites’ continued contributions
to a growing corpus of favored epigraphy. Mu‘izz al-Din and/or the intel-
ligentsia in his service evidently implemented the small epigraphic corpus
associable with the Shansabanis in these annexed territories – and also
farther eastward throughout the Indus’s alluvia. Here, with re-inscription
being the Ghazna Shansabanis’ principal mode of imperial consolidation,
the architecture they did patronize, as well as these buildings’ epigraphic
programs, both become that much more significant (see Chapter 6).
A closer, more granular perspective on the architectural interventions
discussed throughout this chapter yields additional, valuable observations.
Overall, the Ghazna Shansabanis’ re-inscriptions took place in royal and
palatial contexts: these encompassed the probable renovation of the “Palace

Figure 5.23 Lashkari Bazar, Palais aux Raquettes, ruins, exterior, north
entrance. © Photograph Josephine Powell c. 1960. Harvard Fine Arts Library,
of Racquets” close to Bust (Figure 5.23); and the distinctive interventions

at the Ghazna palace (Figures 5.2–5.7), as well as at Lashkari Bazar’s South
Palace (Figures 5.14–5.22). All of these projects most likely would have
been motivated by the intended utilization of these complexes – at the very
least as “stages” for state ceremonial, if not also as residences.
Although there was no specific intervention attributable to the
Shansabanis within the South Palace’s Audience Hall I at Lashkari Bazar
(Figures 5.24–5.28), we recall that abutting it immediately to the south
was the exquisite Oratory F (Figure 5.20), which bore nearly unequiv-
ocal marks of Shansabani-period renovation. Hall I as the architectur-
ally dominant part of the overall complex would have been extremely
important for Mu‘izz al-Din, the new ruler following in the footsteps of
the Yamini–Ghaznavids. Considered by D. Schlumberger to be “without
any possible doubt, the principal place of pageantry, the great audience
chamber of the royal residence [i.e. the South Palace],” the Hall also
initiated the kilometer-long, north–south central axis of the complex as a

Figure 5.24 Lashkari Bazar, South Palace, Audience Hall I (north end of palace). From Schlumberger 1952: pl. XXX.
14/09/21 5:41 PM
Figure 5.25 Lashkari Bazar, South Palace, Audience Hall I, detail of

stucco revetments (photograph c. 1950s). © Courtesy of Musée des Arts
Asiatiques-Guimet, Paris.
whole (Figure 5.14).51 It thereby exemplified and made real the ceremonial
power that the Ghaznavid sultans had wielded as part of Perso-Islamic
kingly comportment, which helped them to retain command of an empire
with both westward and eastward ambitions. The Hall’s gateway loca-
tion over the vast complex of the South Palace (Figure 5.14); its lavish
decoration (Figures 5.25 and 5.26), including a noteworthy epigraphic
program and the famous frescoes of courtly figures in procession; and its
commanding vistas over the Helmand (Figures 5.27 and 5.28) all surely
instantiated an even grander and encompassing notion of kingship than
thus far experienced by Mu‘izz al-Din at the court of his uncle at Bamiyan,
and that of his older brother at Firuzkuh – and we recall that he had spent
a year in Sistan at the court of Malik Shams al-Din Nasri (see above in
this chapter).52 Mu‘izz al-Din’s conquests of Ghazna and Lashkari Bazar,
then, personified the proverbial shoes he would have to fill if he also had
aspirations of extensive dominions radiating in various directions from
Zabul–Zamindawar (maps, pp. xv and xvi).
All of these indications of a transition toward a court-based and pos-
sibly more sedentist lifestyle for the newly titled Shansabani sultan of
Ghazna collectively create a stark contrast to the absence of palatial archi-
tecture at Firuzkuh. Such a divergence is not unprecedented, however: at

268 Iran to India
Figure 5.26 Lashkari Bazar, South Palace, Audience Hall I, detail of paintings.
From Sourdel-Thomine 1978: pl. 9a.
Bamiyan, the Shansabani lineage headed by Fakhr al-Din Mas‘ud (r. ah

540–558/1145–63 ce) seemed to be cleaving toward palatial living, sign-
aling an adaptive accommodation of the localized cultural and political
economy of their immediate environment (cf. Chapter 3). At Ghazna also,
then, the newly established Shansabani lineage appeared to be adopting
the prevalent modes of kingship and the most effective modes of its actu-
ation, which they de facto had inherited from their Yamini–Ghaznavid
predecessors.53
The processes that appear to be evidenced in the architectural traces
of the Shansabani conquest of the Ghaznavids’ erstwhile territories
should not, however, be considered instantaneous or unidirectional.
While Shansabani renovations and modifications of the Yaminis’ palatial

Figure 5.27 Lashkari Bazar, South (or Grand) palace, ruins, looking south along
Helmand River. © Photograph Josephine Powell c. 1960. Harvard Fine Arts
complexes – some undertaken on a grand scale – could be telling indi-

cations of the newcomers’ occupation of these complexes, we would do
well to bear in mind that any shift toward palatial habitation could have
been sporadic, and ultimately would have been gradual: after all, ten years
can be a momentous period of change in the life of an individual, even as
it is virtually invisible in historical time. Perhaps most importantly, any
shift from “tent to palace” did not have to be wholesale or irreversible.
As captured in the concept of the nomad–urban continuum, nomadic–
pastoralist and otherwise transhumant groups became more or less seden-
tist depending on prevalent circumstances and other contingencies (cf.
Introduction and Chapter 2).
In the face of a discernibly growing tendency toward palace dwelling
among the Ghazna Shansabanis and the possible waning of seasonal
migrations in distance and/or frequency, the advantageous “cultural
capital” of nomadism was surely not forgotten. For example, the greater

270 Iran to India
Figure 5.28 Lashkari Bazar, ruins, general view: Central Palace, Pigeon House,
and North Palace, seen from South Palace site looking north along Helmand
River. © Photograph Josephine Powell c. 1960. Harvard Fine Arts Library,
ease with which nomadic and transhumant groups conceptually and

physically understood large swaths of contiguous cultural geographies,
changing topographies, and varying climates; and their ability to traverse
these expanses efficiently and en masse, were all acquired skills that were
extremely valuable in following the geo-historical momentum of their
empire-building Yamini predecessors.54 And these conceptual skills would
be not only useful but fundamental for the sustenance of the Shansabanis
of Ghazna, as explored further in the next chapter.
Notes
1. Cf. Durand-Guédy 2013b: 328–335; Peacock 2015: 143, 166–172; see also
Chapter 3.

2. See Juzjani vol. I, 1963: 405 (trans. vol. I, 1881: 489); Inaba 2013: 77–78 and
passim; Allegranzi 2014: 100–101. See also Volume II.
3. Juzjani vol. I, 1963: 346, 353 (trans. vol. I, 1881: 357, 369–371).
4. Cf. Bosworth 1994: 398–399; O’Neal 2013: 58–59; Chapter 1, above.
5. Cf. Bosworth [1977] 1992: 124–126; Biran 2005: 48–52. The last author noted
that the years of the Oghüz in Khurasan is “the middle period of Qara Khitai
history (1144–77) … [and] certainly the least documented one,” so that more
is known of the conflicts between the Oghüz and the Saljuqs, whose unrav-
eling permitted the nomads to roam unhindered in some areas, including
Ghazna, until their routing by the Shansabanis.
6. Juzjani vol. I, 1963: 396 (trans. vol. I, 1881: 448–449).
7. As noted by Bosworth ([1977] 1992: 111–114), the eastern reaches of the Perso-
Islamic world during the twelfth–thirteenth centuries, particularly the regions
of Afghanistan and parts of Central Asia, received comparatively little atten-
tion in the surviving works of Arabic- and even Persian-language authors, the
latter being “more concerned with such pressing questions as the break-up of
the Great Seljuq empire … and the menaces from Central Asia of the Qara
Khitai and the Khwarazm-shahs.” Thus, Ibn al-Athir and Juzjani provide a
rather skeletal chronology for the events leading to the Shansabani capture
of Ghazna (ah 569/1173–1174 ce) and Gardiz (ah 570/1174–1175 ce), and the
crowning of Shihab al-Din as Sultan Mu‘izz al-Din at Ghazna. Cf. Juzjani
vol. I, 1963: 357–358, 396 (trans. Vol. I, 1881: 376–377, 449); Bosworth [1977]
1992: 125; Edwards 2015: 96; O’Neal 2016a; Chapter 1.
8. The earlier scholarly summation of Mu‘izz al-Din’s status at Ghazna as one
of “gilded vassalage” (Scerrato 1962: 265) has gradually shifted, so that Flood
(2009a: 89ff.) defined “the apogee of Ghurid power” as an unusual arrange-
ment of “the brothers rul[ing] in a condominium” (but cf. H. A. Khan [2016:
24] describing Mu‘izz al-Din as “the second in command of the empire,
who ruled in the name of his elder brother and regent”). As we have seen
throughout this book, however, the centrifugal fragmentation of patrimony
was a common and even usual tendency among many nomadic–pastoralist
societies. Cf. also Introduction & infra in this chapter’s concluding section.
9. See Bosworth [1963] 2015: 36; Planhol [2000] 2012. Ball (2020) noted that
the Rutbils politically co-existed with the Kabul-based Turk-shahis, but their
relationship is still to be clarified. See also Bombaci 1957: 248–250; Bombaci
1966: 4ff.; Lewis 1995: esp. 115ff.; Chapter 3.
10. E.g. Glatzer 1983; Glatzer and Casimir 1983; Barth 1961; Beck 1983; Garthwaite
1983.
11. Such was the case with the Basseri within the Khamseh confederacy, brought
together by a charismatic and enterprising merchant ensuring safe caravan
passage by allying five (Ar. khamsa) tribes under his aegis (cf. Barth 1961:
72, 86–90); and the late eighteenth-century Afsharid mobilization of various
Turkic groups and their eventual confederation as the Qashqai to confront
and negotiate later Qajar domination (Beck 1983: 287–289).
12. Cf. Garthwaite 1983: 314–321.
13. Cf. esp. O’Neal’s (2020: 210–211) analysis of Mu‘izz al-Din’s silver coinage
issued at Ghazna between ah 569 and 581 (1173–1185 ce), where he is named

272 Iran to India
‫ – السلطان االعظم‬effectively appropriating his older brother’s title while refer-

ring to him with none. See also Chapters 3 and 4.
14. For brief descriptions see Ball 2008: 190–193, 240–246 (including Bust); for
additional maps, plans, and bibliography cf. Ball 2019: Nos. 149, 358, 685; and
for a summation of the extant evidence, see Thomas 2018: 59–60, 125–127,
132–135.
15. Upon being pursued by Chinggiz Khan in c.1220, the Khwarazm-shah Jalal
al-Din’s (r. ah 617–628/1220–1231 ce) recourse to Ghazna – his appanage,
where he rallied troops – drew the Mongols there, where he defeated advance
Mongol forces. There was destruction at Ghazna due to the confrontation,
including the killing of its inhabitants and destruction of its jami‘; but ulti-
mately the Khwarazm-shah fled toward the Indus and the Mongols annihi-
lated his forces on its banks in late 1221 (Jackson 2017: 80, 159, 165). Thus, it
is unclear whether Ghazna would have received “more savage treatment than
the vengeful Ghurid ruler ‘Ala’ al-Din Husayn Jahan-suz” had visited upon
the city in c. ah 544/1150 ce (ibid., 180). But even after Chinggiz Khan looked
to new conquests beyond Afghanistan, the remaining bands of Negüderi or
Qara’unas – former contingents of Mongol forces – held the entire territorial
swath from Zabulistan (including Ghazna) eastward into the upper Indus
valleys in their apparently lawless grasp until the early fourteenth century,
making it a “no man’s land.” Cf. ibid., 148, 183, 195, 197; also Bernardini 2020.
16. Prior to the 1950s and the beginning of Italian surveys and excavations in the
Ghazna region, and the French archaeological activity in the vicinity of Bust-
Lashkari Bazar, A. Godard had done initial photographic documentation at
Ghazna in the 1920s, as seen in DAFA archives at the Musée Guimet, Paris
(access kindly granted by M. J. Ghèsquiere, July 2018). J-C. Hackin’s archive
of the 1930s–1940s provided photographs of Bust and its environs: cf. Lashkari
Bazar (1978): planches 42, 46. See also Ball, Bordeaux et al. 2019: 403.
17. For the two episodes of fiery destruction at Lashkari Bazar, see Schlumberger
1978: esp. 31, 37, 41; Sourdel-Thomine 1978: 21, 29, 36, 39, 54; Ball and Fischer
2019: 473; cf. also infra in main text. For the post-Mongol abandonment of
Bust, see Allen 1988: 55, 56.
18. See supra in notes for Lashkari Bazar. For signs of destruction at Ghazna, see
supra in notes for Mongol and post-Mongol impacts; also Bombaci 1966: 5ff.;
Scerrato 1959: 52; Scerrato 1962: 265–266; Giunta 2005: 474ff.
19. See above in main text, and Chapters 1 and 3 for southward expansion by the
Shansabanis of Firuzkuh, after the burning and destruction of Ghazna and
Lashkari Bazar by ‘Ala’ al-Din Jahan-suz (ah 544/1150 ce).
20. Scerrato 1959: 49–52; also Thomas 2018: 126–127.
21. The initial association of the palace and Mas‘ud III began with Bombaci’s
study of a marble fragment found in the modern ziyarat on the western
perimeter of the palace mound: the fragment consisted of the top portion
of a pointed arch with semi-circular lobes within, carrying an inscription
of Q. II: 256 (the Throne Verse) along with the kunya and laqab of the
sultan (cf. Bombaci 1966: 33ff., 9–20). Bombaci already questioned whether
“the palace was really built by Mas‘ud III, or was only readjusted by him.”
Recent re-examinations of the excavation records indicate that the palace’s

irregular quadrangle itself may have been a repurposed caravanserai or other

structure; thereafter, it underwent at least five phases of modification, the
penultimate one being the addition of a large royal oratory on the western
perimeter (approximately where the modern shrine stood): Giunta 2010b:
123–125; Giunta 2020: 164ff.; Allegranzi 2014: 111–112; Allegranzi 2019a, vol. I:
54–56; Allegranzi 2020a; Ball and Fischer 2019: 475–477; Laviola 2020: 28–29;
and Appendix, Afghanistan V: 2.
22. Some of these finds had already been inventoried in the 1950s Italian
Archaeological Missions to Ghazna, e.g., Scerrato 1959: fig.18. See also Giunta
2005: 479–480 and fig. 5; Artusi 2009: 121–122 (quoting Scerrato), 128; Laviola
2018. See also Appendix, Afghanistan V: 4.
23. The marble dado panels recovered from the palace’s main entrance area have
been interpreted as part of a decorative and epigraphic program specific to
this part of the complex; cf. Appendix, Afghanistan V: 1. They are stylistically
related to the dadoes from the central court, and generally dated to the same
period (Appendix, Afghanistan V: 3–4). See Flury 1925: 74; Giunta 2010b:
125.
24. See esp. Giunta 2010b: 126–127; Allegranzi 2020a.
25. Cf. esp. Bombaci 1966: 19–29; also Giunta 2005: 478ff.; Giunta 2010b:
126–127. O’Kane (2009: 22–23 and n. 72) followed Bombaci’s initial study
of the dado slabs and characterized them as a (fabricated) genealogy of the
Ghaznavid sultans beginning in pre-Islamic Iran, referencing Shah-nama.
Recent re-examination of the slabs, and the recovery of others that had been
repurposed as tomb markers, reveal other possibilities: there was likely more
than one poetic composition comprising the inscriptional bands, and the
genres of both the lengthy mathnavi and the panegyric qasida were likely rep-
resented. Notably, references to Islamic juridical and theological ideas indi-
cate that at least some of the subject matter was religious and/or political in
nature. Cf. Allegranzi 2015: 26ff; Allegranzi 2019a, vol. I:145–167; Allegranzi
2020a.
26. Scerrato 1962: 265–267.
27. See esp. Allen 1988: 55–58; Allen 1990: 26; Ball and Fischer 2019: 469–471.
28. Juzjani simply noted that “Mu‘izz al-Din subdued the environs of Ghazna,”
and that the following year (ah 570/1175 ce) he conquered Gardiz. Cf. Juzjani
vol. I, 1963: 396 (trans. vol. I, 1881: 449).
29. Although hesitant to consider this urban configuration as a “peculiarity” of
Khurasan, Rante (2015: 17–20) brought attention to this overall layout in
the development of several settlements throughout the vast region, ranging
from Marv through Nishapur and dating from the Achaemenid through late
Sasanian periods (first–seventh centuries; cf. also Collinet 2015: 127–128 and
fig. 3; Laleh et al., 2015: 118; Ball, Glenn et al., 2019: 272–283). With the Arab
conquests of Central Asia during the eighth century, further quarters (Ar.
rabad) emerged to distribute the settled and incoming populations, with the
Arabs usually within the enclaves. For early Islamic Afghanistan, see Franke
2015: esp. 81–84 and figs. 5, 27; Ball 2019: e.g., Nos. 99, 522, 874, 1006.
30. See Sourdel-Thomine 1978: 63ff.; Allen 1988: 58–60; Ball and Fischer 2019:
474, 544; Ball 2019: No. 149.

274 Iran to India
31. See esp. Ball 2019: No. 149 for elevations of the partly underground structure.
As we have seen in Chapters 2–4, pre-modern textual sources were less than
specific about the structures encountered on military campaigns – and in
the case of the Ghaznavid campaigns, formulaic mentions of “idol-break-
ing” occluded rather than laid bare the details of engagement with the built
environment.
32. Cf. Patel 2004: 54 and n. 24, 108–109, pls. 44–45.
33. For early stepwells in Gujarat – an area that had been pivotal for both
Mahmud Ghaznavi in ah 416/1025 ce, and Mu‘izz al-Din about 150 years
later (cf. Chapter 6), see, inter alia, Jain-Neubauer 1981: 19–21 (and figs.).
A parallel numismatic example of cultural practices moving west from the
Indian subcontinent are jitals, one minted with the stamp of Nimruz: the
location of the mint remains ambiguous, however, as the term denoted a
region rather than a specific place as of the eleventh century; meanwhile, the
city name of Zaranj was replaced by shahr-i Sijistān by the tenth century, even
as Nimruz continued to be a region. By the thirteenth–fourteenth centuries,
Nimruz had come to replace shahr-i Sijistan as a mint name. Thus, the ques-
tion remains whether Nimruz and shahr-i Sijistan were effectively the names
for Zaranj, changing over time (Bosworth 1994: 37, 367, 426, 441). Although
Nimruz/Sistan was considered “the most westerly mint to adopt the jital” (R.
and M. Tye 1995: 50 and Nos. 123, 124), O’Neal (pers. comm., July 2020)
concluded that another example with the mint name of Marv would indicate
the slightly farther western and also northern reach of the jital type. Cf. Nicol
2005: 176; and Volume II.
34. The ambiguity of the structure’s raison d’être cannot be clarified by the remnant
of the construction text, which followed the Qur’anic inscription(s) on the
left side of the arch’s east face (cf. infra in main text, and notes). Although
this fragmentary text contained the expected phrase “… (he who) brought
to completion this … in the year …,” the referent for the definite article was
unclear, except possibly a final ta marbuta. Sourdel-Thomine’s (1978: 65)
suggestion of qubba would certainly imply a memorial function for this arch,
but she proposed this reading as tenuous. See Appendix, Afghanistan II.
35. Sourdel-Thomine (1978: 63) had assumed that the arch formed the entrance
to a large building. After re-surveying the site, however, Allen (1988: 59–60,
fig. 2) proposed that the arch could not have been a building entrance: not
only were both faces of the arch decorated; additionally, insufficient space
around the extant monument made it impossible for a mosque of commen-
surate proportions to have been constructed there. See also Ball and Fischer
2019: 474.
36. The publication (Sourdel-Thomine 1978: 64–65) notes Q. II: 121 as the cited
verse; upon verifying the epigraphy and the author’s translation, however, the
correct citation would be Q. II: 127 – surely a simple typographical error.
37. Sourdel-Thomine 1978: 66.
38. Allen 1988: 59–60.
39. Blair 1992. The verses have been recorded at Qasr Kharana, Jordan and dated
ah 92/710 ce; and on the minaret of Aleppo’s jami‘ in floriated Kufic, datable
to ah 483/1090 ce (the verses’ naskh rendition on Delhi’s ‘Ala’-i Darwaza is

dated ah 711/1311 ce). See esp. Dodd and Khairallah 1981: 2:5. Also Appendix,
Afghanistan II.
40. Schlumberger disagreed with dating portions of the South Palace to
Shansabani presence, e.g., Hall I’s oratory (see below in main text), which
he placed during Khwarazm-shahi occupation c.1215 (noted by Sourdel-
Thomine 1978: 39). In his review of the three-volume publication Lashkari
Bazar (appearing at last in 1978, after Schlumberger’s death), A. D. H. Bivar
(1980) also disagreed with Sourdel-Thomine’s attribution of Shansabani dates
to many parts of the sprawling site. Notably, “in a treatment to be published
elsewhere” the reviewer planned to “query … the ascription of the [Bust]
Arch to Ghurid times” (Bivar 1980: 386); to my knowledge his proposed
re-dating of the monument to the reign of Mahmud Ghaznavi has not yet
been published. Dani (2000: 559) also dated the “magnificently decorated
arch” to the eleventh century without further argumentation. Due to the
current lack of access to the site, and Sourdel-Thomine’s analysis being the
most thorough to date, her argumentation is broadly followed here (cf. infra
in main text).
41. Cf. Sourdel-Thomine 1978: 42–45, 63–68. We have already discussed (cf.
supra in main text) the monumental arch inside Bust’s lower enclave (Figures
5.9, 5.12 and 5.13), a newly built rather than renovated structure to which she
assigned a later twelfth-century date.
42. Schlumberger 1978: 20.
43. For the dating of the Lashkari Bazar renovations principally to ah 55x/1156–
1165 ce, cf. Sourdel-Thomine 1978: 10, 21, 29, 39, 54, 57; also Appendix,
Afghanistan VIII: 2. For Ghazna, see esp. Scerrato 1959: 42–52; Scerrato 1962:
267.
44. At Ghazna, some marble fragments datable to the Ghaznavid occupation of
the site also carried simple Arabic sayings. See Giunta 2010b: 125; Allegranzi
2015: 27–28. These have to be contextualized, however, within the much
longer, sophisticated Persian poetic verses probably gracing the entire interior
surface of the palace’s large central court. Cf. Appendix, Afghanistan V: 3–5;
VIII: 1.
45. Compare the similar configuration of a small oratory ensconced within a large
public space at the Saljuq waystation of Robat Sharaf (near Sarakhs, eastern
Iran) – cf. Chapter 1.
46. Ball (2020) applied this description to both Ghaznavid and Shansabani
architectural decoration, distinguishing it from the more restrained decora-
tion on Saljuq-patronized buildings. Cf. Sourdel-Thomine 1978: 42, 44–45;
Appendix, Afghanistan VIII: 4.
47. See Dodd and Khairallah 1981: 2:9–15.
48. As also noted in Chapter 4 (cf. notes), verses 1–3 of the Sura survive only
in Cairo: at Mashhad al-Juyushi (ah 477/1085 ce); and the citadel of Salaª
al-Din (ah 578/1183–1184 ce), and nowhere else in the Islamicate regions. Cf.
also Chapter 6; and Appendix, Pakistan II: 1.
49. See Appendix II, Pakistan III; and Chapter 6.
50. See esp. O’Neal 2016b; O’Neal 2020; Chapter 1. It is noteworthy that,
despite the Ghaznavid minting of jitals at Ghazna and Bust from the reign of

276 Iran to India
Sabuktigin through that of Khusrau Malik, there are no known jitals attribut-
able to Mu‘izz al-Din from these mints (the farthest western ones being from
Taliqan and Kurraman) (map, p. xv). Although, there were rare copper issues,
which however diverged considerably from the jital type: cf. R. and M. Tye
1995: 41–49, 54 and, e.g., Nos. 83, 85, 87, 88, 89e1, 90, 95, 98, 101, 104, 105e1,
106, 107e1, 108, 112, 115, 190, 191.
51. Cf. Schlumberger 1952: 258–267; Schlumberger 1978: 38–41, 61–64 (my
trans.); Appendix II, Afghanistan VIII: 3.
52. It is noteworthy that at Zaranj, Mu‘izz al-Din would have witnessed the
special status accorded to intellectual dignitaries, who were in fact courted to
lengthen their stays by the Maliks of Sistan, in part to expound on points of
religio-juridical import before an assembly. Such was the case with Juzjani’s
own grandfather Minhaj al-Din ‘Uthman, as well as Ahwad al-Din Bukhari.
See Juzjani vol. I, 1963: 277–278, 396 (trans. vol. I, 1881: 189–191, 447);
Bosworth 1994: 398–399.
53. Among Mu‘izz al-Din’s other possible adoptions from his Ghaznavid pre-
cursors might have been his “collecti[on of] a corps of expensively trained
slaves to make his writ run and his treasures and territories safe from clannish
co-sharers” (Habib 1992: 7). From among the thousands of bandagan-i Turk
Mu‘izz al-Din likely purchased, he had select favorites – e.g., Taj al-Din
Yildiz, Nasir al-Din Qabacha, Qutb al-Din Aibek, Baha’ al-Din Tughril –
all of whom he came to regard as his own children. Cf. also Chapter 2, and
Volume II.
54. In his analysis of the military advantages possessed by nomadic groups, Irons
(2003) employed the concept of “cultural capital” to encapsulate the modes
of organization and skills that developed as a result of specific activities (e.g.,
livestock raiding). In describing the Mongols, Burbank and Cooper (2010:
4) observed their possession of “the technological advantages of nomadic
societies – above all, a mobile, largely self-sufficient, and hardy military –
[and] capacious notions of an imperial society …” (cf. also Chapters 1, and 6).
These studies are also extremely useful in pointing to the possible conceptual
advantages emerging even from varying degrees of nomadic pastoralism. See
also Barfield 1993: 144; Thomas 2018: 25–26.

CHAPTER 6
Encountering the Many “Indias”
I n addition to the architectural traces already attributed to the Ghazna

Shansabanis and analyzed in Chapter 5, several more survive along the
Indus’s shores and in its contiguous hinterlands, in what are the western
reaches of the Indic sphere. I place five architectural sites within the period
of their active presence there, viz. c. ah 575–600/1180–1205 ce – years that
also witnessed their transformative and enduring successes farther east in
the north Indian plains, analyzed in detail in Volume II. As was the case
in the foregoing chapters of the present volume, previous scholars’ work
in the region under examination provides a solid foundation for analysis,
and it is supplemented here with my own architectural documentation.
Admittedly, the survival of historical data is always partial and random,
and it is possible that even among the surviving remains not all have been
correctly identified; new “discoveries” still expand the evidentiary corpus
(as seen below). By gathering the available data and positing new identi-
fications, plausible observations can be made based on what has managed
to reach the moment of analysis. At the very least, I propose overarching
characteristics and patterns that serve as points of departure for further
study.
The first group, already attributed to the patronage of members of the
Shansabani forces, consists of three funerary structures around the eco-
nomically, politically, and spiritually important city of Multan (ancient
Mulasthana) (map, p. xv). A funerary ribat near the village of Kabirwala
(c. ah 576/1180 ce) (Figures 6.16–6.23) lies about 35 km northeast of
the city; based on its foundation inscription, it has been identified as
commissioned by the Shansabani-affiliated sipahsalar ‘Ali ibn Karmakh,
who was the wali of Multan at the time.1 The structure locally identified
as the tomb of Sadan Shahid (Figures 6.24–6.28), located southwest of
Multan near Muzaffargarh, had been documented and also dated to the
late twelfth century, based on comparative stylistic analysis. In 2011, Abdul
Rehman and Talib Hussain published a tomb in the southeastern vicin-
ity of Multan attributed to a certain Ahmad Kabir (Figures 6.29–6.32);
the tomb’s inscribed (in Arabic) date of ah 600/1203–1204 ce, as well
as its overall form and decorative iconography, have both conclusively
associated it with the Shansabani presence in the middle Indus region.

278 Iran to India
The rediscovery of this dated structure and its formal and iconographic
parallels with the tomb of Sadan Shahid have bolstered the latter’s dating
to the late twelfth century.
At the necropolis of Lal Mara Sharif (Figure 6.37), located 250 km
northwest of Multan toward Dera Ismail Khan but still along the Indus’s
alluvia, four domical structures made of brick clearly distinguish them-
selves from their surroundings due to their glistening, glazed-tile exterior
and interior decorative programs. To date, they have been studied most
intensively by Taj Ali and Holly Edwards, who attributed them either to
the early eleventh-century Ghaznavid activities in the region, or to the
second quarter of the thirteenth century, following the Mongol incur-
sions.2 However, based on our preceding analyses of the Shansabanis’
pre-imperial and imperial architectural traces in Afghanistan and the
revised dating of some of these structures (cf. Chapters 2–4), I consider
the four tombs as part of the Shansabani affiliates’ architectural patronage
in the area.
Finally, two more tombs are located 450 km southwest of Multan and
in the vicinity of modern Sukkur (Figures 6.33 and 6.34) – itself at the
crossroads linking Panjab, Baluchistan, and Sindh (see below). They had
been documented and briefly discussed by A. N. Khan (1988), and more
extensively by M. Kevran (1996b), who each offered varying dates ranging
from the tenth through early thirteenth centuries, again based on stylistic
analysis and comparison with earlier structures.3 But I propose that these
tombs also be dated to the last two decades of the twelfth century, and thus
considered part of the corpus of Shansabani-period funerary structures in
the middle Indus region (see full discussion below). Collectively, then, all
three of these architectural groups are invaluable sources for the nature of
empire-building undertaken by Mu‘izz al-Din and his dependents as they
followed the geo-political momentum radiating eastward from Ghazna.
As will be discussed throughout this chapter, the structures attribut-
able to Shansabani patronage in the Indus region evince an astound-
ing variety in formal styles and also materials of construction. The one
uniting factor among them, however, is the overall westward qibla ori-
entation of the buildings with mihrabs (only the tombs of Sadan Shahid
and Ahmad Kabir lacked mihrabs altogether). Precise coordinates are not
available for these complexes but, based on the observations in Chapter 4
regarding pre-modern qibla orientations, it is fair to say that the qiblas of
Shansabani-patronized buildings at the western edges of the Indic world –
that is, Sindh and along the upper Indus’s shores – were probably neither
accurate nor consistent.
The general westerly direction of the surviving mihrabs, however, would
indicate the qibla preferred by the Hanafis. This madhhab had remained
dominant among the later Ghaznavid sultans despite Sultan Mahmud’s
leaning toward the Shafi‘iyya.4 Although, it should be noted that no

Encountering the Many “Indias” 279
unequivocal indication of Hanafi affiliation has been recovered from the

extensive epigraphic program of the Ghazna palace. Nevertheless, Juzjani
claimed that upon being crowned sultan of Ghazna, Mu‘izz al-Din,
“in conformity with the people of the city of Ghazna and its environs,
accepted the madhhab of the exalted Imam [Abu Hanifa].” The westerly
qibla orientations of the Ghazna Shansabanis’ architectural traces along
the Indus, then, would have been consistent with their new avowal of
Hanafi juridico-legal affiliation.5
Indus Bound
Nothing like the disappointment of an elder, the inertia of historical

momentum, and a little good luck to act as impetuses for ambitious
actions with reverberations well beyond one’s own lifetime! This con-
vergence of factors could partially serve to explain – as plausibly as any
other – the tireless campaigns of territorial expansion and consolidation
that the recently crowned Sultan Mu‘izz al-Din undertook within four
to five years after taking possession of Ghazna and the erstwhile Yamini
territories of southern Afghanistan. Mu‘izz al-Din’s eastward campaigns
increasingly differentiated his new Shansabani lineage based at Ghazna
from his agnates at Firuzkuh and Bamiyan. Furthermore – and, of course,
unbeknownst to Mu‘izz al-Din at the time – the spectacular victories of his
military forces, extending ever deeper into the Indian subcontinent, were
to have irreversible longue durée ramifications for the flow of people, ideas,
and things between the larger Indic and Iranian cultural worlds.
I argued in Chapter 5 that the Ghazna Shansabanis’ expansionist ambi-
tions, and the underlying motivations behind the numerous and ambi-
tious campaigns early on in Mu‘izz al-Din’s reign, did not appear out of
nowhere. Rather, they came as the inextricable inheritance of conquering
Ghazna, which was integrated into networks of commerce and communi-
cation that were strongly oriented toward India, and geo-politically poised
for eastward imperialism. In important recent works, Waleed Ziad and
Ali Anooshahr have both compellingly proposed that, ever since the first
decade of Mahmud Ghaznavi’s reign (ah 389–421/999–1030 ce), “the
relationship of the Ghaznavid court with Hindu India was not simply
one of predatory violence, but imperial over-lordship”.6 While the major-
ity of historical authors patronized by the Ghaznavid rulers emphasized
the near annual campaigns into northern India by Mahmud and Mas‘ud
I (r. ah 421–432/1030–1041 ce) as raids for plunder and destruction of
“idol-houses” or temples, other important evidentiary sources indicate
that the Ghaznavids’ aims were more encompassing, their victories aimed
at gaining rights of sovereignty in India (see the following sections of this
chapter).

280 Iran to India
At least a few of the reported incidents of temple destruction were

formulaic or exaggerated, and may be only the most legible result of larger
intentions.7 It has been argued, for example, that Mahmud’s infamous
Somnath campaign of ah 415/1025 ce was much more than a raid to strip
a temple of its riches: in reality, it was likely a failed attempt at coastal
access for an otherwise land-locked empire; gaining entry into the Indian
Ocean networks via the coast of Gujarat – the hinge linking the western
and eastern halves of the Indian Ocean world – would have been a most
profitable maritime debut.8 Upon occupying Ghazna, then, Sultan Mu‘izz
al-Din came into a capital whose sovereigns had purported to be both
Perso-Islamic as well as Indian kings (cf. Volume II).
Inheriting the geo-political momentum of Ghazna goes far in explain-
ing the newly crowned Shansabani sultan’s military sorties and campaigns,
undertaken almost immediately upon conquering the Ghaznavid capital
and its dependent areas. The greater facility in commanding large swaths
of geography and climate – part of the “cultural capital” resulting from
nomadic pastoralism and/or transhumance in lifestyle and ancestry (cf.
Chapters 1, 2, and 5, above) – surely served as a continuing advantage for
the sultan. Indeed, we saw in Chapter 1 that Mu‘izz al-Din’s very first
campaign was to the physically and culturally remote region of Gujarat
in ah 574/1178 ce, only four or five years after occupying Ghazna. This
initial campaign becomes all the more remarkable when we recall that
Mahmud himself had campaigned but once in Gujarat, and that only
toward the end of his reign, with no other Ghaznavid sultan campaigning
there again.9 In this particular instance, the new Shansabani sultan of
Ghazna might have overestimated the advantages afforded by his experi-
ence in nomadism and/or transhumance and its “cultural capital,” and the
concomitant apprehension of contiguous geographies and climates. But he
was quick to recover and plot strategies that ultimately led to remarkable
successes in India, which were extremely consequential for centuries to
come (see Volume II).
Inherited Landscapes, Re-inscribed
The Ghaznavids, particularly the sultans Mahmud and Mas‘ud I, had cer-
tainly crafted an imperial Perso-Islamic presence westward in Khurasan, but
they had also blazed pathways eastward from the plains of Ghazna, leaving
traces in the architectural–cultural zones of the western Indic world as they
proceeded to the duab of northern India. The inertia of this economic and
geo-political momentum, which Mu‘izz al-Din maintained after his con-
quest of Ghazna, presented him with yet more inherited landscapes (cf.
Chapters 1 and 2): a Perso-Islamic imperial “infrastructure” – consisting
of royal or elite-patronized mosques and other public institutions – that

had been laid by his Ghaznavid precursors.10 Much of the Shansabani

activity in the same areas, along with their own imperial marks, are fruit-
fully conceived, then, as re-inscriptions of already existing socio-political
assets within their own jurisdiction, for example, by means of substantial
endowments (Ar.-Pers. waqf) to pre-existing architectural complexes and
the institutions they housed. The larger territorial acquisitions were dele-
gated to allied military leaders and their dependents.11
Chapter 2 explored the pre-imperial Shansabanis’ extensive adaptation
and reuse of the landscapes they inherited throughout historical Ghur.
Given the early Shansabanis’ greater transhumance and/or nomadism
during the eleventh and early twelfth centuries, new construction of
purpose-built, monumental architecture was neither needed nor possible –
the latter due to the lack of sufficient surplus means to patronize laborers
skilled in building traditions beyond the nomad–urban continuum. Things
did, of course, shift considerably during the intervening three-quarters of a
century or so, both in lifestyle as well as economy: by the early decades of
the 1100s, it is probable that the Shansabanis were politico-economically
surpassing other Ghuris, and their upward trajectory necessitated increas-
ingly monumental architectural patronage. If the minar of Da Qadi has
been correctly identified as that of Qal‘a-yi Zarmurgh and accurately dated
to the first quarter of the twelfth century (Chapter 2), we could surmise
that, by this time, the Shansabanis had access to builders with some famil-
iarity with Persianate traditions, and their ability to implement these on
a localized level. Indeed, by the 1170s, large-scale construction had been
undertaken by the Shansabani lineages at Firuzkuh and Bamiyan, and to
a somewhat lesser extent by the newly crowned Sultan Mu‘izz al-Din at
ex-Ghaznavid sites, all of which surpassed the scale and quality of their
own earlier efforts and placed them firmly within the courtly trends of the
larger cosmopolitan Persianate world (cf. Chapters 3–5).
It is in fact Mu‘izz al-Din and the Ghazna Shansabanis who stand out
as we explore farther eastward. Not only were they the protagonists of
these successful campaigns ever deeper into the Indic cultural world, they
had also distinguished their particular mode of building empire as one
primarily based on the re-inscription of pre-existing environments rather
than the creation of new ones, as amply evidenced at Ghazna and Lashkari
Bazar, and probably also Bust (see Chapter 5). As further demonstrated
by the Shansabanis of Bamiyan also, those pre-imperial tendencies toward
adaptation of inherited built forms and landscapes were not entirely super-
seded by the patronage of new buildings. But in the Indus-dependent
areas and beyond, rather than only physical reuse or repurposing of inher-
ited structures, all remnants of past empires and their attendant societal
assets – particularly those of the Yamini–Ghaznavids – were meaningfully
re-inscribed, layered with a Shansabani mark in addition to that of their
immediate predecessors.

282 Iran to India
The cities of Lahore and Multan serve as plausible examples of the

Ghazna Shansabanis’ practices of re-inscription, as they entered the western
corridors of the Indic sphere. Sultan Mahmud Ghaznavi had annexed
Lahore probably quite early in his reign, near the beginning of his regular
campaigns into the north Indian duab. The poet Fakhr-i Mudabbir (d.
1228; cf. Chapter 1), patronized by Iltutmish (r. ah 606–633/1210–1236 ce)
at Delhi and writing during the early thirteenth century, made mention of
Mahmud’s commemoration of an Indian victory in ah 411/1019–1020 ce
with the construction of a towering minar at Lahore.12 Like its later parallels
at Ghazna and Jam-Firuzkuh, the Lahore minar was probably associated
with a nearby, contemporaneous mosque, possibly even the one identified
in the lower strata of excavations within Lahore Fort.13 Given the city’s
immense importance as the point of entry into and retreat from the north
Indian plains, it was directly administered by deputies with military and
legal authority, specially appointed by the Ghaznavid court throughout the
reigns of Mahmud and Mas‘ud I.14 And we have already seen that Lahore
was the refuge of the last two Ghaznavid rulers after the Shansabanis
wrested Ghazna in ah 569/1173–1174 ce. Thus, this city was certainly well
marked with an imperial Ghaznavid presence, and its continued strategic
importance would make it equally valuable for Mu‘izz al-Din’s ambitions.
Despite the absence of identifiable Shansabani-patronized architectural
traces, the city could nonetheless have been appropriated and re-inscribed
with Shansabani presence, for example, by means of continuing endow-
ments to already existing mosques, madrasas, and other “civic” institutions.
This was precisely the Shansabani method of appropriating the long
prosperous city of Multan, at the nodal crossroads linking Panjab with
Sindh and Zamindawar, and destinations beyond in virtually all direc-
tions. Since at least the seventh century and probably earlier, this city
had also been renowned for its temple to Aditya, a Brahmanical–Hindu
solar deity, serving as an exemplar of the historical nexus intertwining
commerce and pilgrimage.15 In the late tenth century, Multan’s Isma‘ili
da‘i Halam ibn Shaiban finally destroyed the temple on the orders of the
Fatimid khalifa al-Mu‘izz li-Din Allah (r. ah 341–364/952–975 ce) and
replaced it with a mosque. This new foundation was apparently in close
proximity to another mosque, probably the masjid-i jami‘ founded upon
Muhammad ibn Qasim’s conquest of the city c. ah 95/714 ce (with later
renovations and enlargements, of course).16 The Shi‘a-affiliated Isma‘ili
presence in Sindh, whose missionary activity in the region had begun
perhaps half a century before,17 had cultivated strong trading networks
between the Mediterranean and the western coasts of the subcontinent
via the Indus, far beyond the regional overland routes that had converged
on the Aditya Temple. It was surely the profitable Isma‘ili-directed com-
merce that prevented Mahmud from definitively eradicating them on
his campaign there in ah 395/1005–1006 ce, despite their heretical status

within Sunni Islam. According to Juzjani, Isma‘ili communities remained

in Multan until Mu‘izz al-Din’s campaign there in ah 571/1175–1176 ce,
when their presence finally came to an end (see also above).
After taking Multan and its vicinity from the last of the Isma‘ilis,
Mu‘izz al-Din evidently re-inscribed the area within the domains of the
Ghazna Shansabanis: he made substantial endowments to various insti-
tutions, including the incomes of two villages for defraying the overall
operating costs of the city’s masjid-i jami‘, as well as further provisions
for the mosque and its mu‘adhin, and five salaried instructors for madra-
sas. While it is possible that Mu‘izz al-Din constructed a mosque and/or
madrasas to which he made these additional endowments, it seems more
likely that such institutions – particularly the masjid-i jami‘ – would have
already existed in a prosperous and important city like Multan, which also
had a long Islamic affiliation. Mu‘izz al-Din’s endowments, moreover,
remained in place at least for the following two centuries, being recorded
in the surviving correspondence of the mid-fourteenth century by the
Tughluq-deputed governor of the city.18
Thus, the Shansabani sultan of Ghazna made a substantial mark on
Multan and possibly also Lahore, even without patronizing new and
monumental architecture but rather by appropriating their important
institutions via endowments intended to exist in perpetuity, re-inscribing
the cities as having entered within Shansabani domains. It is precisely
the “blended-in” quality of Shansabani architectural traces that provides
strong continuity with their discretely adaptive, region-specific modes
of gaining control of the physical and cultural environments that they
encountered during the early decades of their ascendancy, both in their
homeland of Ghur and beyond.19
The clustering of surviving Shansabani-period sites along the middle
Indus – traversed by the northeast–southwest corridors spanning the his-
torically significant cities of Multan and Lahore and reaching beyond
into Baluchistan (discussed subsequently in this chapter) – is certainly
convenient for scholarly analysis. But an exclusive focus on these archi-
tectural traces would not elucidate the architectural and cultural geogra-
phies the Ghazna Shansabanis encountered on their other campaigns, for
example, the northeastward foray to Peshawar and historical Gandhara
and Swat in ah 575/1179–1180 ce, or the southeasterly push into Sindh in
ah 578/1182 ce (cf. further below). Built remains of Shansabani presence in
these regions may yet come to light; but even without securely identified
traces of their successful campaigns there, we can reasonably suppose that
the Shansabanis marked or otherwise re-inscribed their conquered tracts,
which were most often commercially and strategically important urban
centers and their associated environs. Preceding a concentrated treatment
of the Shansabani-period tombs within the Indus-dependent areas, a brief
excursion into Gandhara–Swat and Sindh–Baluchistan–Makran helps

284 Iran to India
us to understand the varied cultural geographies, or the “many Indias,”

which the Shansabanis encountered, further bearing out their adaptive
and region-specific approach to building and consolidating a confederate
empire.
Gandhara and Swat

Even the westernmost reaches of the Indic cultural world – now within
modern Pakistan and eastern Afghanistan – were far from an unvarying
or uniform cultural geography. Rather, I propose that the region com-
prised at least two discernible architectural–cultural areas: the first, in
the north interconnecting upper Panjab through historical Gandhara and
Swat; the second, in the south encompassing Sindh, the Makran coast,
and eastern Baluchistan. The differing – though not divergent, but rather
complementary – historical trajectories of these respective areas from
the early historic period onward, and their distinct cultural productions,
necessitate their separate treatment. Albeit brief, this examination allows
us to trace the varied landscapes the Ghazna Shansabanis came upon
during their first forays eastward.
Mu‘izz al-Din’s campaign to Peshawar (ah 575/1179–1180 ce) – and
most likely beyond into the surrounding valleys – brought him and the
Ghazna Shansabanis face to face with a region that was home to various
flourishing communities, whose identities crisscrossed confessional and
occupational axes of belonging. Major Buddhist monasteries (Figure 6.1)
Figure 6.1 Buddhist monastic complex, Takht-i Bahi, Khyber-Pakhtunkhwa

Province, Pakistan. © Photograph Alka Patel 1997.

and other religio-economic establishments such as temples (see below)

had come up along important overland routes of communication during
the first millennium ce, helping travelers to traverse the difficult terrain
connecting Eurasia and the Indian subcontinent.20 It was precisely the eco-
nomic vibrancy of the region that had attracted Muslim military adven-
turers of the eighth–ninth centuries, who staked claims of control over
pockets within the region. Similarly, the emergence of the Hindu-shahi
rulers (c. 843–1026 ce) of Udabhandapura (near modern Attock on the
Indus) – followers mainly of the Shaivite sects of Brahmanical Hinduism
(cf. Chapter 1) – also relied on the economic networks of commerce and
pilgrimage throughout ancient Gandhara and Swat to build their empire.21
The ascendancy of the Ghaznavids in Zabulistan by the end of the
tenth century, and their eastward expansion shortly thereafter, brought
the Gandhara–Swat region within the sights of Sultan Mahmud not only
for its economic offerings, but also due to its function as one of the prin-
cipal conduits into the north Indian plains. The region’s midway position
between the Khyber Pass and northern India ultimately rendered it an
important waystation for the Ghaznavids, where there was little resist-
ance to their entry, especially after the final defeat of the Hindu-shahi
Anandapala in ah 400/1010 ce: certainly, this is the impression gleaned
from the Ghaznavid sources’ scanty mention of campaigns into Swat, and
equally so from their architectural traces. Udegram’s terraced Rajagira
mosque and its small surrounding settlement (Figure 6.2) are among the
few known Ghaznavid-period remnants in the region. Commanding a
spectacular view of the valley below, the mosque of “Gandharan” masonry
(Figure 6.3) – slabs of local schist in mortar – formed a continuation with
the building methods employed in the region’s Buddhist monasteries since
the early centuries of the common era.22
Toward the southwest leading to the Panjab, at the southern limits of
historical Gandhara, Ghaznavid-period authors enumerated several mili-
tary confrontations specifically between Sultan Mahmud and Hindu-shahi
rulers. Although no Ghaznavid-period architectural traces have defini-
tively been identified in the area,23 a distinct and enduring style of temple
architecture – spanning the sixth–seventh through eleventh centuries and
largely developed via Hindu-shahi patronage – was clearly in evidence. The
Ghazna Shansabanis’ campaign to Peshawar and beyond, then, brought
Mu‘izz al-Din and his forces into the erstwhile Hindu-shahi territories and
surely included encounters with their architectural remains.
The scholarship of Michael Meister24 has been seminal in understand-
ing the architectural developments in the Salt and Hissar mountain ranges,
spanning the modern Pakistani provinces of eastern Khyber-Pakhtunkhwa
and West Panjab – the latter’s historical extent reaching southward to the
middle Indus around Multan. Meister convincingly posited the distinc-
tiveness of this regional school of temple architecture, which he called

Figure 6.2 Rajagira Mosque, Udegram, Khyber-Pakhtunkhwa Province, Pakistan. © Photograph
Alka Patel 1997.
14/09/21 5:41 PM
Gandhara–Nagara, differentiating it
from parallel eastern developments in
northern and southern India.25
In arguing for the distinction of the
Gandhara–Nagara school of temple
architecture, Meister identified its
three fundamental sources, which
were a convergence of indigenous and
imported conventions. Straddling both
categories was the Buddhist architec-
ture and iconography (c. 100 bce–500
ce) of historical Gandhara, which
encompassed localized interpretations
and applications of Greco-Roman
forms and iconographies. These
Buddhist remains spanned northwest
Pakistan through eastern Afghanistan,
largely the area where Gandhara–
Nagara temples are found. The influx
of north Indian Nagara forms (fifth–
seventh centuries ce) was equally
evident, though locally interpreted
based on the available building mate-
rials and also distinctive rituals (see Figure 6.3 Rajagira Mosque, Udegram, mihrab.
below). Finally, elements from the © Photograph Alka Patel 1997.
temple architecture of Kashmir were
also traceable in the regional temples’
architectural fabric.26 Although no architectural treatises codifying the
Gandhara–Nagara school have yet been found, Meister’s comprehensive
analyses of the surviving buildings have elucidated the developments in
the regional style’s forms and iconographies over time: decorative elements
deriving from the Hellenistic past of the area, for example, continued
to play important iconographic roles in Gandhara–Nagara decoration
(Figure 6.4). Additionally, the distinctive and ubiquitous trefoil arch, pos-
sibly deriving from Kashmiri architectural forms, was also a characteristic
feature of Gandhara–Nagara temples (Figures 6.4–6.6).27
The surviving temple complexes provide invaluable insights, including
possibly new functions and rituals that distinguished them from their
eastern counterparts. Unlike the north Indian examples, Gandhara–
Nagara temples were constructed of the local volcanic stone (kanjur) hewn
into blocks, permitting multi-storied interiors (Figure 6.6). The upper
stories were usually single chambers, roofed with domes resting on stepped
squinches (Figure 6.7). Access to the upper floors was provided by dark,
narrow inclined corridors that were dimly illuminated and ventilated by

Figure 6.4 Temple, Kalar, West Panjab, Pakistan, exterior façade. © Photo by Alka Patel 1997.
14/09/21 5:41 PM
apertures disguised in the temples’

exterior surface decoration (Figure
6.8). Given these temples’ notable
formal (and structural) diver-
gences from their north Indian
cousins, they surely served differ-
ent uses – some possibly defensive,
as suggested by Meister.28 They
certainly housed localized rituals
that seemed to distinguish Shaivite
and other Brahmanical–Hindu
religious practices in the region
from those of the north Indian
plains and elsewhere. Only further
archaeological excavations and
studies will clarify the contours
of these important local devel-
opments. Indeed, the abilities of
architectural traditions to meet
localized ritual and other needs
over time was further borne out
into the twelfth century: as argued
later in this chapter, the identifia-
ble Shansabani architectural traces
in the vicinity of the great city of Figure 6.5 Temple, Mari Indus, West Panjab,
Multan were in themselves further Pakistan. © Photograph Alka Patel 1997.
developments of the Gandhara–
Nagara architectural canon.
Sindh, Baluchistan, and the Makran Coast

In ah 578/1182 ce, the Ghazna Shansabani campaign to the renowned
port of Dibul – now accepted to be the modern archaeological site of
Banbhore29 (Figure 6.9) – brought Mu‘izz al-Din and his forces face to face
with the commerce-driven, extremely prosperous, cosmopolitan region of
coastal Sindh, which also extended to meet Gujarat toward the east, and
to the west the Makran coast and its hinterlands in southern Baluchistan.30
Merchandise and riches poured into and out of the area with equal vigor
via the Indus, the mighty yet capricious river that served as an artery of the
vast Indian Ocean world as it linked global maritime and overland routes.
Juzjani’s description of Mu‘izz al-Din’s campaign as one “captur[ing]
the whole of that seaside country” swept into broad strokes the many small
and large entrepôts dedicated to the bustling flow of commerce, as well as
the substantial urban centers that rose and fell depending on the Indus’s

290 Iran to India
ever-changing course. In fact, just a

century after the Shansabani cam-
paign (and of course unanticipated by
Mu‘izz al-Din at the time), Banbhore–
Dibul itself would experience a dra-
matic contraction of trade traffic as the
Indus shifted several kilometers to the
east, displacing Banbhore–Dibul in
favor of Lahori Bandar as the princi-
pal riverine port.31 Nevertheless, at the
time of the Ghazna Shansabanis’ entry
into this world, Banbhore–Dibul and
nearby commercial centers and ports
still served as proverbial lynchpins
along the Indus, connecting world-
wide Indian Ocean littorals with the
overland travel routes spanning the
Eurasian landmass.
In large part thanks to the sus-
tained commercial and cultural con-
nections of the continuous coastline
of Gujarat, Sindh, and Makran, the
settlement of West Asian populations
Figure 6.6 Large temple, Ambh Sharif, West here and farther inland is documenta-
Panjab, Pakistan. © Photograph Alka Patel 1997. ble since the early historic era.32 Rather
than disrupting this flow of people
and ideas, the emergence of Islam as an identifiable set of socio-religious
affiliations and practices easily fit within the networks interconnecting
essentially the entire “Old World.” The immigration and integration of
Muslim communities into the polities of the western coasts of the Indian
subcontinent, then, was far from a novelty by the late twelfth century.
Polities ruled by Muslim elites in this region – or with substantial pres-
ences of Muslims – were the result not only of the earlier ‘Umayyad
through Ghaznavid incursions (cf. Chapters 2 and 3), they were also sus-
tained by local peoples with long histories of conversion to Islam and
Islamization.33
To date, scholarly works on coastal northwestern India have – perhaps
understandably –focused on the prosperous mercantile and other elites
inhabiting the cosmopolitan urban centers dotting the region (cf. above
and Bibliography). However, surviving material indices remind us of the
participation of other, understudied populations in creating the regional
dynamic of an inclusive cosmopolitanism, wherein multiple socio-religious
mores and cultural forms coexisted. The late tenth-century influx of the
Saljuq confederacies into the Iranian world, and their gradual extension

Figure 6.7 Temple, Nandana, West Panjab, Pakistan, interior, top story.
Figure 6.8 Large temple, Ambh Sharif, exterior, detail of shikhara

(curvilinear superstructure) surface decoration. © Photograph Alka Patel 1997.

292 Iran to India
Figure 6.9 Mosque remains, Dibul-Banbhore, Sindh, Pakistan. © Photograph

Alka Patel 1997.
as far west as Anatolia by the twelfth century, also had reverberations into
coastal India: Throughout the tenth through twelfth centuries, various
nomadic–pastoralist groups identified as Baluch were displaced from their
encampments, pastures, and fields within the ambit of the Caspian Sea,
thereby undertaking long-distance overland migrations into the eventually
eponymous region of Baluchistan.34 Thus, a historically more accurate idea
of cosmopolitanism – at least as gleaned from this well-trafficked region
of South Asia during the twelfth century – should encompass and ensue
from the cultural expressions of both urban-centered elites as well as other
groups along the nomad–urban continuum.
Architectural traces of the wide range of peoples, their economies and
lifeways in Sindh and its contiguous areas reinforce the need for this
inclusive notion of cosmopolitanism, to which Mu‘izz al-Din and the
Ghazna Shansabanis would have contributed as they forayed into the
region. Modest single-room, brick-constructed but elegantly decorated
funerary structures such as the tomb attributed to Muhammad ibn Harun
at Lasbela (historical Armabil), generally dated to the eleventh century
(Figure 6.10),35 evince clear typological and formal dialogues with the vir-
tually unstudied, probably earlier or contemporaneous funerary structures
scattered throughout Baluchistan – for example, around Lasbela itself,
Kharan, Kalat, Panjgur, and Jalawar. Abundant terracotta plaques with
geometric and floral motifs, as well as figural scenes of agriculture and
herding, clad almost the entirety of these structures’ exterior surfaces.36
Small-scale cultivation and animal husbandry were not only life-sustaining

Figure 6.10 Tomb attributed to Muhammad ibn Harun, Lasbela, Baluchistan,

Pakistan. From Edwards 2006, fig. 3.
and predominant among the various Baluch and other groups and con-
federacies settled or transhumant in the region, supporting the attribution
of these mortuary structures to them. Indeed, the selective adaptations
and reinterpretations of Islamic orthodoxy among nomadic pastoralists
and other transhumant groups permitted the continuation of non-Islamic
customs, which could be the basis for figuration on these nominally
“Islamic” structures.37
Notably, the lavish decoration of the above-described commemora-
tions was reduced to a series of “textile-like band[s] of carved brickwork”
on the tomb of Muhammad ibn Harun – predictably without figuration –
and relegated to the upper parts of its elevation. The tomb’s interior was
also plain except for a pointed-arch mihrab within a rectangular frame of
zigzag bricks.38 Taken altogether, it appears that the practice of cladding
built surfaces with decorated terracotta plaques was a wide-ranging indige-
nous practice throughout Makran–Baluchistan and Sindh, documentable
at least since the mid-first millennium ce on pre-Islamic remains,39 and
adapted to the architectural needs – quite possibly of a funerary nature – of
settling or transitory populations such as the Baluch nomadic pastoralists
and others. Ultimately, terracotta cladding was also integrated into the
Islamic tombs of the region, but without the “non-Islamic” iconogra-
phies. Indeed, it is difficult to sustain a strict division between figural/
pre- or non-Islamic, and non-figural/Islamic decoration, so that figural
and non-figural plaques may have been made concurrently for different
patrons.

294 Iran to India
2. Ruins of the Mosque of Thamban Wari

__________________________
1. Sketch of Location of Jam Jaskars Goth

Fortification and Congregational Mosque
__________________________
3. Plan of the Mosque

__________________________
6. Epigraphy Fragments from

the Congregational Mosque
__________________________
5. A Column of the Mosque

__________________________
4. Reconstruction of Mosque
__________________________
Figure 6.11 Congregational Mosque and Thamban Wari Masjid, Jam Jaskars
Goth, Sindh, Pakistan, plans, elevations, drawings. After Kevran 1996: figs. 9–14.
These buildings in the localized traditions – some attributable to the

Baluch and/or other nomadic pastoralists, and some to local saints and
heroes – were not the only built forms marking the landscape. The larger
projects patronized by elites in the region subscribed to distinct building
practices. For example, only 7 km southwest of Banbhore–Dibul – and
well within the circuit of the Ghazna Shansabani campaign of ah 578/1182
ce – lay Jam Jaskars Goth, one of the many large and small commercial
centers, fortified warehouses, and waystations punctuating the shores of the
Indus over time as it changed course toward the east.40 This site’s extensive
remains included a quadrangular fort (7,900 m2), a large congregational
mosque (3,000 m2), and a smaller but more complete mosque (about 1,300
m2); altogether “these monuments [were] not those of a village, but rather

those of an important agglom-

eration.”41 The exquisitely dec-
orated smaller mosque here,
locally known as the Thamban
Wari masjid (“the mosque of
columns”) (Figure 6.11), exem-
plifies the style and material of
construction favored by the elite
strata of the polities spanning
Gujarat, Sindh, Makran, and
southern Baluchistan; it was a
mode of building that co-existed
with the above-described local-
ized traditions.
Jam Jaskars Goth’s Thamban
Wari masjid was paved with
brick, though the rest of its
structural fabric – including
the eponymous columns – were
hewn from locally available
stone. More intact and roughly
contemporaneous structures at
the site of Bhadreshvar, along
the Kachh coast of Gujarat,
provide comparable examples of
flat-roofed mosques (Figure 6.12) Figure 6.12 Chhoti masjid, Bhadreshvar, Kachch
and tombs, some with corbeled District, Gujarat, India, interior, column. © Photograph
“domes” (Figure 6.13). Thamban Alka Patel 2001.
Wari masjid has received only
scanty publication thus far, but it is apparent that its structural compo-
nents were carved with a confident expertise, the multi-shafted columns
consisting of square, circular, and eight- or sixteen-sided sections. Their
decorative programs seemed to draw from an established iconography,
including motifs of diamonds in series; crisply rendered but still lush
foliage; keyhole niches; and most notably the stylized but recognizable
overflowing pot motif (Skt. purnaghata) well known from the decorative
iconographies of the region’s temple and other architectures (Figure 6.14).
Given the masjid’s stylistic continuity with the surviving stone temples
of the area, the colonial-era archaeologist Henry Cousens assumed that it
“had been constructed from the materials of a Hindu temple … [and] the
Muhammadans chiseled out the images from the niches on the pillars.”42
In more recent scholarship, the mosque has been summarily compared
with “the stone temples of the region and the … mosques and tombs
constructed for maritime communities of Muslims”.43 But rather than

296 Iran to India
Figure 6.13 Tomb known as Shrine of Ibrahim, Bhadreshvar, interior facing west.
resulting from a process of local artisans meeting the ritual requirements of

settling Muslim communities on an ad hoc basis, it is important to under-
stand the structure as part of an established set of architectural practices
prevalent in the lower Indus region, to be distinguished from those else-
where in the western reaches of the Indic world (as discussed earlier in this
chapter). Discerning the specific set of traditions and practices from which
the building emerged provides not only a more complete understanding
of the Indic sphere’s extremely variegated western areas, but also a material
glimpse into the Ghazna Shansabanis’ encounter with yet another land-
scape of elite “orthodoxies” (cf. Chapter 4), which was quite distinct from
their expanded homelands in what is modern Afghanistan.
It would be accurate to state that, by the late twelfth century, South
Asia’s Islamic ritual structures in the broadest terms constituted archi-
tectural typologies that had been seamlessly added to the repertories of
identifiable, regionally based building traditions and practices.44 The sty-
listic and structural features of Jam Jaskars Goth’s Thamban Wari masjid
are consistent with what the great scholar of temple architecture M. A.
Dhaky (1927–2016) called the Maru–Gurjara style.45 This architectural
canon originated in the contiguous region of Gujarat–Rajasthan to the
east, now in modern India, and extended westward into Sindh, develop-
ing in tandem with Chaulukya supremacy in the overall area during the
mid-tenth through late thirteenth centuries.46 The Maru–Gurjara archi-
tectural canon consisted of a prescribed set of building practices with
micro-regional developments over time. Arguably, practitioners of the

Figure 6.14 Temple, Bhodesar, Nagarparkar District, Sindh Province, Pakistan, exterior base of shikhara. © Courtesy of Fatima Quraishi.
14/09/21 5:41 PM
298 Iran to India
Maru–Gurjara style were equipped to erect a great variety of buildings,

ranging from mosques, tomb structures, elaborate sarcophagi, towering
cenotaphs, and likely non-religious architecture.47 Active patronage of the
Maru–Gurjara style can be discerned well into the eighteenth century,
spanning the areas from eastern Rajasthan through westernmost Makran,
along with the contiguous inland areas in southern Baluchistan.48
At least one Maru–Gurjara architectural treatise, the Jayaprrcha of the
early twelfth century, prescribed methods for identifying suitable terrain
for building mosques (Skt. rahmana-prasada), and detailed the configura-
tions of their plans, elevations, and decorative programs.49 The relation-
ship between formulated architectural knowledge and its application was a
complex and reciprocal one across time and space, comprising experience
through building innovation; accumulation and codification of prescribed
principles; and eventual modification of these prescriptions with more
building experience.50 The survival of very few treatises into modern times,
moreover, makes tracing this dialogical process tenuous at best. Rather,
a much more complete understanding of architectural innovations over
time – including the creation of new building types and decorative ico-
nographies – has been provided by the surviving buildings themselves.
The Jam Jaskars Goth Thamban Wari masjid and other Islamic ritual
structures in contiguous areas are effectively the visible “records” of these
intertwined processes: The structural and iconographic innovations, real-
ized as Muslim patrons commissioned locally trained artisans to meet
their architectural requirements, were then incorporated within the larger
Maru–Gurjara canon. This canon was a responsive rather than an ossified
point of reference, which enabled craftspeople to innovate within it, as
they rose to the twin challenges of expanding architectural typologies as
well as meeting the individual preferences of old and new patrons alike.
Thamban Wari masjid has been tentatively dated to the eleventh–
twelfth centuries.51 But without a date-bearing inscription or enough sur-
viving epigraphy for paleographic analysis,52 the structure’s more definitive
dating will depend on thorough stylistic and iconographic comparison
with the region’s more securely dated architectural remains. In the mean-
time, however, it is worthwhile noting that this building of the Maru–
Gurjara style – and/or others throughout the many urban settlements
scattered in the alluvia of the lower Indus – could have been commis-
sioned by Mu‘izz al-Din or his affiliates among the Ghazna Shansabanis.
Since their pre-imperial days in inherited architectural–cultural land-
scapes, the Shansabanis’ wont had been to adapt existing resources toward
immediate ends (see esp. Chapter 2). Now on the brink of a transregional
politico-economic ascendancy, their modus of building empire was emerg-
ing: rather than imposing an architectural ideal on their newly conquered
territories, they patronized and effectively helped to develop the local
building practices already prevalent in their regions of imperial activity.

Tombs along the Indus
Among the innumerable architectural remains from South Asia’s past – and
generally in large swaths of the world’s architectural heritage – defensive
and sacral structures have survived in greater numbers than their resi-
dential (particularly non-palatial) and other non-religious counterparts.
Over time, buildings with religious and defensive functions have tended
to be made of more durable materials such as baked brick, stone, or
a combination thereof (cf. esp. Chapter 2). Moreover, these structures’
continuing usefulness as sites of devotion or military importance – and in
fact a combination of these functions in the riba†, as discussed below – has
earned them some form of investment in their well-being by the various
communities surrounding them through the centuries. Certainly, these
broad observations go some way in explaining the apparent preponderance
of Shansabani-period tombs as traces of their forays eastward into the
western reaches of the Indic world.
We have seen that, though Mu‘izz al-Din maintained the eastward
momentum of his Yamini–Ghaznavid predecessors, his actual mode of
building empire deviated considerably from theirs in several respects; at
least two of these deserve mention here. First, since the Ghaznavids had
already laid – both proverbially and literally – the foundations of imperial
control, Mu‘izz al-Din and his followers frequently re-inscribed existing
structures and even large landscapes upon annexation of the Ghaznavids’
territories, as discussed above. Further, the Shansabani architectural traces
that do survive – some already identified, others newly proposed here –
were all of a funerary nature. An examination of possible reasons for this
proclivity toward funerary architecture among the Ghazna Shansabanis
indicates the continuation of some pre-imperial practices, as well as
changes wrought in these practices as their Shansabani-affiliated patrons
encountered the funerary traditions along the Indus’ shores.
Quite independently of the Shansabanis, the burgeoning of Sufi net-
works linking South and Central Asia at least since the eleventh century,
and likely earlier, led to the proliferation of – and increasing focus on –
funerary architecture in the Indus region. But it is worth considering that
an earlier Shansabani tendency toward expending resources on funerary
sites and structures, discussed in Chapter 3, was also a factor in the pre-
ponderance of Shansabani tombs (discussed below). These structures were
primarily of non-imperial patronage, likely commissioned by the trusted
military leaders who – in addition to Mu‘izz al-Din himself – actually con-
stituted the Shansabani advance eastward from Ghazna. The structures’
smaller scale, and their emergence largely from local rather than imported
architectural practices, have contributed to the inherent difficulties in
attributing them to Shansabani presence in the region.
The nexus between commerce and pilgrimage, so prevalent along the

300 Iran to India
Indus’ alluvia, only grew more powerful after the Isma‘ili destruction of
Multan’s Aditya Temple during the later tenth century. Although this act
might have disrupted the overland pilgrimage traffic to the city – quite
possibly from as far afield as Zamindawar (cf. above and Chapter 3) –
commercial benefits overall actually increased: the active strengthening of
maritime and transcontinental ties thanks to the Isma‘ili da‘wa, established
in the region since the mid-ninth century, did serve Fatimid political ends
and also provided a platform for Isma‘ili proselytism in India. But the
Isma‘ili combination of religious zeal and commercial acuity – supported
by the Fatimid khalifa – only increased the prosperity, renown, and reach
of the area’s commercial networks far beyond the overland routes of before.
It could be said that a significant architectural–cultural consequence
of the commercial prosperity all along the Indus’ dependent cities and
hinterlands – driven in substantial part by Shi‘a-affiliated Isma‘ilis – was
an increasing sacralization of the region’s landscape. The supposedly het-
erodox status of Shi‘ism within broader Islam – and even more so of
Isma‘ilism – perhaps acted as a natural magnet for the Sufi silsilas prolif-
erating in western through central Eurasia, and making their way toward
new horizons.53 The ongoing contests among political powers to command
the middle Indus region during the late eleventh through twelfth centuries
did not alter Multan’s pivotal position in the commerce–pilgrimage nexus
encompassing the Indus’s shores.
In fact, this city in particular was emerging as a “city of tombs”: it had
come to be the perfect place of convergence for prominent and obscure
Sufi orders, each collectively founding their khanaqahs and immortalizing
their deceased pirs with shrines large and small, some of which com-
manded wide devotion for centuries to come (Figure 6.15).54 Thus, Multan
and other Indus cities, along with their hinterlands, came to be sacralized
landscapes punctuated by commemorations to deceased spiritual masters.
The charisma these spiritual men often possessed meant that they were not
always able or indeed willing to eschew the very worldly ambit of polit-
ical intrigue.55 By the late twelfth century, the accumulation of funerary
sites throughout the middle Indus region could well have subtly urged a
Shansabani emphasis on tombs or other types of funerary structures as
well.
It should also be noted that the proclivity toward investing resources
in commemorative sites had already been visible among the Shansabanis
themselves during their early expansions beyond historical Ghur in the
later 1150s. In Chapter 3 we analyzed the material traces of Shansabani
presence in the area of al-Rukhkhaj–Zamindawar, specifically at
Tiginabad/Old Qandahar (map, p. xvi). In c. ah 553/1158 ce, ‘Ala’ al-Din
Husain Jahan-suz had again disrupted the Ghaznavids’ corridors of
movement between Ghazna and Zamindawar (and ultimately Sistan and
Herat): following on the heels of his humiliating defeat of Bahram Shah

Figure 6.15 Multan, West Panjab, Pakistan, tomb of Sheikh Yusuf Gardizi.
in ah 544/1150 ce and his vengeful destruction of Ghazna and Lashkari

Bazar, he wrested Tiginabad/Old Qandahar and its hinterlands from the
Ghaznavids. Excavated Shansabani-attributed traces at the site appear to
be funereal, in the form of two large tombs that simultaneously adhered
to and also deviated from orthodox Islamic burial practices. It is well
known that the many alterations of the region’s landscape over the cen-
turies render it extremely difficult to obtain a full view of Shansabani
activity there. Additionally, the caution of archaeologists that assessments
of nomadic societies should not be reduced to “an archaeology of graves”
is a valuable one.56 And yet, the preponderance of funerary buildings
among the Ghazna Shansabanis’ architectural patronage in the middle
Indus region appears to support the idea that commemoration of the
dead had commanded a significant portion of their material and cultural
resources from earlier days. The perception of a predominantly funerary
nature of architectural traces as they made their momentous entrée into
the western reaches of the Indic world, then, may not be entirely the result
of historical accident.
Remarkably, even the relatively few buildings or complexes datable to
the period of Shansabani entry into the Indus region are themselves further
distinguishable in three groups: the funerary riba† of ‘Ali ibn Karmakh
(Figures 6.16) and the other two, smaller single-room tombs, respectively
attributed to Sadan Shahid (Figure 6.24) (near Muzaffargarh, southwest
of Multan) and Ahmad Kabir (near Dunyapur, south of Multan) (Figure
6.29), share architectural features directly traceable to the above-described,

302 Iran to India
distinctive and regionally dominant Gandhara–Nagara style of temple

architecture, as discussed in greater detail below. By contrast, the two
“twin” tombs at Sukkur (Figure 6.33) – about 450 km southwest of Multan
on the northern borders of historical Sindh – possess the “domed-cube”
profile that was extremely common for funerary structures throughout
the Islamic world by this time (see Chapter 4, and below). Important
elements in their decorative programs, however, link them to pre-Islamic
traditions; but, rather than affinities with Gandhara–Nagara temple
architecture as in the Multan riba† and tombs, I suggest that the Sukkur
structures were much more likely to have relied on the iconographies
emerging from the Maru–Gurjara architectural ambit, patronized from
the Gujarat–Rajasthan region in the east and reaching far westward into
Sindh, Baluchistan, and Makran (see above in this chapter). Finally, the
standing tombs at Lal Mara Sharif (Figure 6.37), approximately 250 km
northwest of Multan (40 km south of Dera Ismail Khan) differ notably
from both of the preceding architectural groups: even though all four
buildings here exhibit variations of the “domed cube,” their profiles are
much less attenuated than at Sukuar, and they have extensive glazed-tile
decorations on their exteriors and interiors.
Examining each of these groups in turn serves multiple purposes: such
a focused exploration certainly demonstrates the great architectural “ecu-
menicalism” in the Shansabani-affiliated military leaders’ patronage of
buildings along the Indus’ riverbanks. This reliance on the locally available
skilled labor and material is a palpable continuation of the tendencies that
had been characteristic of all three Shansabani lineages as they expanded
within Afghanistan (cf. Chapters 2–5). Furthermore, the groups of build-
ings themselves furnish a unique glimpse of the coexistence of several
architectural cultures along the westernmost borders of the Indic cultural
sphere, along with the mobility of architectural and iconographic ideas via
the people who communicated them.
The Multan Region

To date, scholarly analyses of the Shansabani-patronized structures sur-
viving in the vicinity of Multan – viz. the riba† of Ali ibn Karmakh and
the two single-room tombs of Sadan Shahid and Ahmad Kabir – have
“dissected” them into their respective “Indus Valley” and “Ghurid” or
“Central Asian” components, with the admirable goal of apprehending
how these buildings came to look and function in the way they did his-
torically. Invariably, these structures’ fascinating epigraphic programs
have been ascribed to the latter categories, seemingly balancing out their
“non-Islamic” decorative components by underscoring the consummately
“Islamic” practice of citing Qur’anic verses within architectural iconogra-
phies.57 By parsing the various contributing elements, however, we may

be missing the specific and even unique factors that converged to create
these buildings’ forms, and which endowed them with ritual and other
functions – albeit still only partially understood.
I propose that the Indus structures ensued from a combination of
modes of manufacture: first, it should be borne in mind that buildings
dedicated to Islamic ritual had probably been incorporated into regionally
produced architectural treatises during the previous one to two centuries,
and perhaps longer. This process of architectural codification, described
above in relation to the prevalence of the Maru–Gurjara style along the
Gujarat–Sindh–Makran coasts and their hinterlands (also further discussed
below), was the result of an already assimilated combination of various
“ingredients.” Forms, iconographies, and configurations, recognizable
across the surviving buildings, had been naturalized as part of an entirely
accepted method of making ritual architecture for Muslim patrons within
this particular architectural culture (cf. also Volume II). By continuing to
dissect the buildings and list their constituent proportions and decorative
details individually, we as scholars not only perform unnecessary work, we
may actually misunderstand the processes by which these buildings came
to look and function in the way they did at the historical moment being
examined, namely, the Shansabani encounter with the western frontiers of
the Indic world.
Second, the overall decoration and especially the epigraphic programs –
the focus here – of these buildings are significant: the specific Qur’anic
verses appearing on the Indus structures evince direct connections to the
Shansabanis’ architectural projects in central Afghanistan, initiated during
the mid-twelfth century. The Qur’anic content of the decorative programs
of the Afghan buildings was apparently being repeated, thus coalescing as
an identifiable corpus on the Indus-area structures built in the wake of the
Ghazna Shansabanis’ successful campaigns eastward. In fact, much of this
content was carried farther into the Shansabani architectural patronage in
the north Indian duab (cf. below in notes and Volume II). Admittedly,
none of the three structures in the Multan vicinity is preserved in its
original state – either due to unchecked deterioration or, in the case of
Ahmad Kabir, renovation. But, despite having a less than complete idea
of their epigraphic programs, it is nonetheless worth observing that the
Indus buildings’ epigraphy had meaningful connections particularly with
the Shansabani-patronized structures farther west.
The Shansabani-affiliated structures near Multan fall within the south-
ern reaches of the Gandhara–Nagara style – discussed previously in this
chapter – manifesting many of its recognizable decorative iconographies,
such as the distinctive trefoil arch (Figures 6.17, 6.27, 6.32). Although few
in number, these Shansabani-period buildings evince a consistency in both
form and iconography to indicate the artisans’ reliance on received archi-
tectural knowledge, rather than ad hoc experimentation. The tombs of

304 Iran to India
Sadan Shahid and Ahmad Kabir

were square, single-room struc-
tures originally capped by domes
atop ogival squinches (Figures
6.26, 6.31). All four of their exte-
rior façades have complementary
decorative programs, under-
pinned by trefoil-arch niches
framing the familiar Islamic dox-
ologies (at Sadan Shahid [Figure
6.27]), or Qur’anic verses (at
Ahmad Kabir [Figure 6.32]; dis-
cussed below). The lowermost
moldings (Figures 6.28, 6.30) of
their exterior elevations carried
floral-arabesque and zoomorphic
elements clearly emerging from
the long-standing pan-Indic,
Buddhist, and Brahmanical
Hindu iconographic reper-
toires, and in the case of Sadan
Shahid also the geometric knot
Figure 6.16 Riba† of Ali ibn Karmakh, Kabirwala, near recognizable from plaited Kufic
Multan, West Panjab, Pakistan, plan. From Edwards 1991: epigraphic programs.58 Both of
fig. 1 (after Mumtaz 1985: fig. 3.3). these tombs additionally have
horizontal and vertical running
bands of Qur’anic inscriptions in naskh (Figures 6.28, 6.30) and historical
inscriptions in Kufic (discussed below). Notably, the tomb of Ahmad Kabir
also carried a Sanskrit donatory inscription in Sharada script on its south-
ern façade. Although dual-language historical inscriptions were known
particularly throughout northern South Asia,59 the presence of Sharada
script calls for further research, as it was apparently limited to the region
of Kashmir as of the early eleventh-century demise of the Hindu-shahis.60
The much larger riba† of ‘Ali ibn Karmakh (Figures 6.16 and 6.17)
followed a well-known architectural typology of the fortified monastic
establishment common in Gandharan and Bactrian Buddhism (see also
Chapter 3). The immense practicality of this architectural typology likely
rendered it a popular building type into later centuries.61 But rather than a
central courtyard bound by rows of cells or rooms on most if not all four
sides, this riba†’s plan (Figure 6.16) consisted of multiple, perpendicular
corridors surrounding what was likely a domed central tomb chamber.
Significantly, the riba†’s well-studied semi-circular mihrab (Figure 6.18) –
whose lavish epigraphic program and other decorative iconography was
centered around the (by now) familiar trefoil arch (Figures 6.19 and

Figure 6.17 Riba† of ‘Ali ibn Karmakh, near Multan. © Photograph Alka
Patel 1998.
6.20)62 – was part of the western north–south corridor, rather than being
ensconced within a defined prayer area. Only more thorough examination
of the riba†’s architectural fabric and ideally its excavation will reveal
whether the built space was reconfigured over time. But the occasional
presence in the walls of carved brick fragments reused as building materials
(Figure 6.21) indicates modifications, which could have altered the space
surrounding the striking mihrab.
Despite its now seemingly auxiliary status within the riba†, the mihrab
serves as a quintessential synopsis of what had surely coalesced into a
well-heeled set of iconographic (and architectural) conventions within
the Gandhara–Nagara school. The ease with which not only pan-Indic
but also locally interpreted iconographic elements have been deployed in
the overall program strongly indicates expertise earned through practice:
for example, the two different but equally successful renditions of the
purnaghata (overflowing pot motif) – schematic on the innermost vertical
epigraphic band (Figure 6.20), but more sculptural and fully rendered as
capitals on the pilasters supporting the springing of the trefoil arch (Figure
6.22); as well as the molding sequence of concave and convex dentils,
fleshy lotus leaves (Skt. padmapatra) and the angular fillet forming the
base of the niche’s hood (Figure 6.23), all indicate a deep familiarity with
iconography and its placement that was appropriate in an Islamic context.
Moreover, the impressive baked-brick renditions of Qur’anic verses in
continuous friezes around the mihrab (Figures 6.24 and 6.25) – not unlike
the larger friezes on the exteriors of the other two Multan-area tombs –
were clearly an addition to the Gandhara–Nagara iconographic repertoire,

306 Iran to India
Figure 6.18 Riba† of Ali ibn Karmakh: interior, mihrab. © Photograph Alka
Patel 1998.
added as Islamic ritual buildings were constructed in the region over the
centuries. Akin to Maru–Gurjara treatises codifying the prescriptions for
making the rahmana-prasada (see above in this chapter), then, it is more
than likely that the Gandhara–Nagara school had also developed just such
a set of accepted conventions for local Islamic buildings, contained in
treatises now lost or yet to be recovered.
The new Shansabani-affiliated patrons of Islamic ritual architecture in
this region marked their territorial conquests not only with their build-
ings, but also by mobilizing locally based skilled laborers for construc-
tion of these projects. It is further probable that, given the distinctive
characteristics that had emerged in the Shansabanis’ earlier buildings in

Figure 6.19 Riba† of Ali ibn Karmakh: interior, framing bands of mihrab, in situ
c. 1980s. From Edwards 1991, pl. IXb.
Afghanistan – explored in previous chapters – their architectural prefer-

ences made some impact on the Gandhara–Nagara region. Although there
are no surviving Ghaznavid-period architectural traces in the middle Indus
area63 with which to compare subsequent constructions – as was the case
in Afghanistan – even by themselves the standing Shansabani-patronized
buildings may provide plausible clues regarding their architectural contri-
butions to the region.
As discussed in Chapters 4 and 5, the epigraphic programs on surviving
Shansabani-patronized complexes in Afghanistan had consisted almost
exclusively of Arabic and partially Persian historical inscriptions, Qur’anic
verses, and doxologies. The Shansabani elites as erstwhile nomads and/or
pastoralists had minimal exposure to the vast and extremely refined literary
cultures of the Persianate world, and possessed a limited conversance with
the Qur’an itself – a limitation that was surely even more pronounced in
Qur’anic interpretation and exegesis. Therefore, with very few exceptions,
the vast majority of Shansabani-patronized architectural inscriptions were
likely selected by the ‘ulama’ and other members of the courtly intel-
ligentsia, who either moved with them into the annexed areas or were
already based there.64 One exception to this overall tendency could have
been a portion of the epigraphy on the main southern façade of Shah-i
Mashhad (c. 1165–1175 ce) (Figure 4.1): here, along with the date and
name of the founder, the original inscription might have included up to
the first eight verses of Surat al-Fath (Q. XLVIII). Given the rarity of these
verses’ occurrences on monuments throughout the Islamic world, I argue

308 Iran to India
Figure 6.20 Riba† of Ali ibn Karmakh: interior, detail of epigraphic bands framing
mihrab. From Edwards 1991, pl. IXa.
that royal preference could have played a decisive role at Shah-i Mashhad
as the Shansabani patron(s) emulated the imperial Ghaznavids through
epigraphy (Figures 4.10 and 4.11).65
Thereafter, the beginning verses of Q. XLVIII again appeared within the
Shansabani-associated renovations at Lashkari Bazar (Figures 5.20–5.22),
which led us to observe that an epigraphic corpus could have been forming

Figure 6.21 Riba† of Ali ibn Karmakh: interior, detail of reused fragment.
within Shansabani architectural patronage.66 The Indus-area buildings lend

support to this observation: The Sura’s first three verses (at least) were again
quoted on the Tomb of Sadan Shahid’s eastern façade (Figures 6.25, 6.28),
forming a frieze that began as a vertical band on the northeast and continued
horizontally, terminating on the southeast of the façade.67 The frame-like
arrangement of Qur’anic text – though different verses – was also prevalent
at the riba† of ‘Ali ibn Karmakh (Figures 6.21 and 6.22), and interpreted as an
iconographic configuration that likely came to form part of the prescribed dec-
oration of Islamic ritual buildings specifically within the Gandhara–Nagara
architectural ambit (cf. above). Further evidence for this coalescing group
of Qur’anic quotations comes from the other small shrine near Multan, the
tomb of Ahmad Kabir (Figures 6.29 and 6.30), where all four verses of Q.
CXII (al-Ikhla‚) appeared on the east and north façades. We recall that this
Sura had made its debut in the Persianate regions of the eastern Islamic world
at Chisht (1160s), reappearing at Lashkari Bazar’s exquisite Oratory F (Figure
5.21), and yet again here along the banks of the Indus.68
Indeed, the architectural projects undertaken by the Ghazna Shansabanis
along the banks of the middle Indus further expanded the epigraphic
corpus of Qur’anic verses that we had seen on Shansabani-patronized
monuments in Afghanistan. The “new” verses appearing along this eastern
frontier of their expansionist campaigns were the following (cf. also
Appendix, Pakistan I–III): on the mihrab of the riba† of ‘Ali ibn Karmakh
are Q. IX: 18–19 and 129 (al-Tauba); on the tomb of Ahmad Kabir’s north
and west façade is Q. I: 1 (al-Fatiha), additionally, on the west façade are

310 Iran to India
Figure 6.22 Riba† of Ali ibn Karmakh: interior, detail of pilaster framing mihrab
niche. © Photograph Alka Patel 1998.
Q. LV: 1–2 (al-Rahman) and Q. CVIII: 1–3 (al-Kauthar), and finally, on

the tomb’s east façade is Q. XLII: 20 (al-Shu’ura).69
Many of these verses were not wholly unusual, also being attested
on buildings in Iran and Transoxiana.70 One exception to this overall
tendency, however, was Q. XLII: 20 (al-Shu’ura) on the southeastern
end of the tomb of Ahmad Kabir’s façade. This Sura was not among the
inscriptions documented by Blair on architecture in Iran and Central
Asia dating from the ninth through the early twelfth century.71 But
intriguingly, Q. XLII: 19–20 were present on the late fifteenth-century
Bara Gumbad mosque in Delhi. Meanwhile, later verses such as 26–27
of Q. LV graced the arched portal of the tomb of Mir Sayyid Bahram at
Kirminiyya (Uzbekistan), which Blair dated c. ah 500/1106 ce.72 But the

Figure 6.23 Riba† of Ali ibn Karmakh: interior, detail of base moldings of mihrab
hood. © Photograph Alka Patel 1998.
Figure 6.24 Tomb of Sadan Shahid, Muzaffargarh, near Multan, West Panjab,
Pakistan: exterior, west façade. © Photograph Alka Patel 1998.
Sura’s earlier verses were instead documented at Delhi’s Qutb Mosque

(Q. LV: 1–14), founded by the Mu‘izzi ghulam Qutb al-Din Aibek c. ah
588/1192 ce after the momentous victory at the second battle of Tara‘in.
Verses 1–12 appeared also on the outer southern entry to the tomb of
Iltutmish (c. ah 630/1233 ce) in the same complex. This tomb also carries

312 Iran to India
Figure 6.25 Tomb of Sadan Shahid: exterior, east façade. © Photograph Alka
Patel 1998.
Figure 6.26 Tomb of Sheikh Sadan Shahid, interior. © Photograph Alka

Patel 1998.
Q. CVIII: 1–3 (al-Kauthar) on the same entry. Finally, even though Q.

CXII: 1–3 (al-Ikhlas) were present at Chisht (cf. Chapter 4), it is also worth
remarking that Q. CXII: 1–4 were carved in relief at Delhi on the north
entry to the tomb of Iltutmish (cf. Volume II).73

This is precisely where we might underscore an added

significance of the Shansabanis’ middle Indus Qur’anic
quotations: like the verses from Surat al-Fath (Q. XLVIII)
traveling with the Shansabani conquests from Shah-i
Mashhad in central Afghanistan to the tomb of Sadan
Shahid along the banks of the Indus, they may provide a
glimpse of the larger body of preferred Qur’anic content
originally appearing on no longer extant or partially pre-
served buildings in Afghanistan. Equally important to note
is that many of these verses continued to appear on struc-
tures patronized specifically by the Ghazna Shansabanis in
north India, reinforcing the idea that there was indeed an
identifiable, enduring corpus of Qur’anic verses through-
out the vast expanse of Shansabani conquests. Figure 6.27 Tomb of
The other two groups of tombs datable to the Shansabani Sadan Shahid: exterior,
presence at the opposite ends of the Indus’s flows, namely detail of decorative niche.
those at Sukkur in Sindh and Lal Mara Sharif in Panjab, © Photograph Alka
appear to have been decorated with pseudo-epigraphic Patel 1998.
designs containing no decipherable mes-
sages (Sukkur), or simple linear elements
(Lal Mara Sharif). Moreover, these two
groups derive from distinct architectural
traditions – perhaps understandably, given
the approximately 800 km separating
them on the Indus’s southern and north-
ern extremes, respectively. Therefore, they
receive separate treatment below.
The Lower Indus

On stylistic grounds, A. N. Khan dated the
two tombs at Sukkur to the late twelfth–
early thirteenth centuries, falling within Figure 6.28 Tomb of Sadan Shahid: north
the timeframe of Mu‘izz al-Din’s cam- half of east façade. © Photograph Alka
paign to Banbhore-Dibul (ah 578/1182 ce) Patel 1998.
and the probable posting of deputies in
Sindh.74 In support of this dating, it should also be noted that the type of
script mimicked in the tombs’ pseudo-epigraphy is stylistically resonant
with the epigraphy on the Firuzkuh Shansabanis’ monumental projects
farther west: the angular treatment of hastae is combined with equally
angular knots, evoking the rendering of the script within the exquisite
epigraphic program at Shah-i Mashhad (cf. Chapter 4). In local tradi-
tion the tombs are attributed to saintly figures, one to a certain Sheikh
Shakarganj and the other to one Khattal al-Din Shah, but neither of these

314 Iran to India
Figure 6.29 Tomb of Ahmad Kabir, Dunyapur, near Multan, West Panjab,
Pakistan, east façade. From Rehman and Hussain 2011, fig. 21.
© Color photograph courtesy of A. Rehman.
Figure 6.30 Tomb of Ahmad Kabir: exterior from northwest. © Courtesy of

A. Rehman.

personages appears in other histor-

ical sources. In the 1980s when H.
Edwards conducted her fieldwork
and documentation, she noted that
the local inhabitants also referred
to the tombs as commemorating
Suhagan (“the bride”) and Duagan
(“the pious woman”).75 The latter
structure was much better preserved
in the 1990s, and serves here as the
focus of analysis relevant to both
(Figures 6.33 and 6.34).
The structures are located only
about 200 meters from each other
and were clearly conceived accord-
ing to the same overall concept:
their profiles were determined
by a cube with a high-drummed
dome, the latter appearing more
prominent atop the cubes’ rela-
tively small volumes below (Figure
6.33). The interiors are devoid of Figure 6.31 Tomb of Ahmad Kabir: interior.
decoration, with the exception From Rehman and Hussain 2011, fig. 20. © Color
of double squinches, pyramids of photograph courtesy of A. Rehman.
dentils punctuating the transition
corners, and a row of diagonally placed bricks at the springing of the
domes in both tombs. Their exteriors, however, are visually striking. In
both cases, the cubical body is divided into three horizontal registers:
the bottom-most sequence of moldings has a practiced and familiar feel,
pierced on four sides by entries and collectively providing the base for
the tall middle section. Here, each of the four sides of the cube is divided
into tall, rectangular blind niches by means of pilasters with purnaghata
capitals – recognizable despite their schematic renditions – upholding
scrolled brackets (Figure 6.34). The resulting rectangular niches serve as
the ground for cut-brick geometric and pseudo-epigraphic decoration.
The cubical body’s corners were conceptually “reinforced” by more
pilasters with the same capitals. This principal portion of the elevation
is balanced by horizontal bands of cut-brick, pseudo-epigraphic decora-
tion, and further geometric bands culminating in a cornice.
Although tombs of saintly figures varied widely, and some grew into
imposing structures with centuries of devotional visitations and dona-
tions,76 the Sukkur structures’ tall presences hearken to a different purpose.
The buildings’ overall proportions and rather lavish exterior decorative
programs – including tall pseudo-epigraphic bands – point to a “worldly”

316 Iran to India
character more befitting tombs of

politico-military actors. Further, the
Shansabanis’ ambivalent engage-
ment with the minutiae of epigraphy
did not preclude the desire for the
presence of script; but surely saintly
and perhaps even erudite personages
would have preferred that the script
convey an actual message, rather than
using it for mere decoration. During
the last decades of the twelfth century,
the need to consolidate territories
already traversed by the Ghaznavids
furnished the circumstances for
architectural commemorations of the
military deputies who were in Mu‘izz
al-Din’s service as he pursued imperial
ambitions eastward toward the Indus
(and beyond).
But the Sukkur tombs are both
geographically and formally quite
distant from the tombs of Sadan
Shahid and Ahmad Kabir, the latter
located about 450 km northeast in the
vicinity of Multan. Given the devel-
opment of the distinctive Gandhara–
Nagara school of architecture
throughout the Panjab into modern
Khyber-Pakhtunkhwa Province, it
may be tempting to attribute all of the
Figure 6.32 Tomb of Ahmad Kabir: exterior recognizable iconographic elements
detail of decorative niche with Quranic inscription. of the Sakkhar tombs – such as the
© Courtesy of A. Rehman. purnaghata capitals – to this style
as the only repository in the region
from which to draw such decoration. But close stylistic analysis – not to
mention an awareness of geographical location – strongly suggest other
sources for the decorative programs of the Sindh structures.
As discussed above, the region of Sindh had long been incorporated
into the ambit of the Maru–Gurjara architectural tradition, radiating
eastward from the areas of Rajasthan–Gujarat in north India. While
many exemplars of this style evince a highly sculptural approach to
architecture, replete with a notable plasticity in both structure and
decoration, more planar and architectonic strains also survived. I have
argued elsewhere that it was precisely this strain that was seminal to

Figure 6.33 Tomb of Duagan, Sukkur, Sindh, Pakistan. © Photograph

Holly Edwards.
creating an architectural language for Islamic ritual buildings in

the region.77
The rendition of the purnaghata capitals can be meaningfully diagnostic
of architectural tradition as well as timespan. By the later tenth century and
onward, the pot overflowing with lush foliage, the purnaghata, had already
reached varying degrees of stylization and even abstraction within the more
architectonic branch of the Maru–Gurjara style. Both temples as well as
the Islamic ritual buildings of Gujarat and Sindh showed progressively
more motif-like renditions of the purnaghata (Figures 6.14, 6.35), reducing
the iconographic concept to its constituent parts and at times embellishing
each of them. At Sukkur (Figure 6.34), this process of abstraction reached
a culmination, probably due to reasons of stylistic interpretation as much

318 Iran to India
Figure 6.34 Tomb of Duagan, Sukkur, exterior, detail of pilasters.

© Photograph Holly Edwards.
Figure 6.35 Chhoti Masjid, Bhadreshvar: interior, detail of pilaster capital.

as the medium of baked brick, which precluded truly plastic treatments

of three-dimensional forms. Furthermore, while pseudo-epigraphic dec-
oration is not known within the Maru–Gurjara ambit, Kufic epigraphy
(Figure 6.36) – the visual source for the angular pseudo-epigraphy at
Sukkur – had become part of the iconography of Islamic ritual buildings
constructed within this tradition. Thus, rather than reflexively attributing
all of the Shansabani-period structures along the Indus’ shores to the

Figure 6.36 Shrine of Ibrahim, Bhadreshvar: foreporch, detail of Kufic inscription

at base of ceiling. © Photograph Alka Patel 2001.
Gandhara–Nagara style as a generic source for their forms and iconogra-

phies, detailed analysis suggests that Shansabani-affiliated patrons of struc-
tures adhered to the architectural conventions that were prevalent in the
respective sites’ larger landscapes. Once again, this adaptive command of
their newly consolidated territories echoed the earliest indications of their
activities throughout the varied cultural geographies of Afghanistan.
The Upper Indus

So far, the compelling architectural evidence for the Ghazna Shansabanis’
continuation of their adaptive command of new territories – an inclusive
modus operandi that relied on local building practices – creates an unu-
sually varied vista of “conquest”: rather than imposing an architectural
signature of pre-designed forms and iconographies, we have seen that
Shansabani-affiliated patrons embraced ways of building that were already
in place, facilitating the further development of these local practices. There
was apparently no “imperial style” that was consistently evoked by the
Ghazna Shansabanis. The absence of such a model, or set of features, with
which scholars might identify architectural remains certainly renders much
more difficult the already challenging process of tracing the transition to
elite status of seasonally transhumant and other obscure, non-elite groups.
But the absence of such restrictions can also permit reconsideration of
architectural remains previously attributed to different dates or patrons.
Indeed, in light of recent scholarship and emerging evidence, it is

320 Iran to India
Figure 6.37 Lal Mara Sharif, near Dera Ismail Khan, West Panjab, Pakistan, tombs at necropolis.
worth reconsidering the dates of the four tombs with mihrabs at Lal
Mara Sharif (Figure 6.37), located about 250 km northwest of Multan
(and about 750–800 km north of Sukkur). As noted by H. Edwards,78
the four tombs can be conceived in two groups, wherein Tombs I and
II have circular, bastion-like towers addorsed on all four corners, while
Tombs III and IV are simple domed cubes; Tomb IV additionally has
an octagonal drum supporting its dome. These formal variations do not,
however, seem to indicate any great difference in dating, as the glazed-tile
decorations and other details largely unify the structures. Collectively,
the Lal Mara buildings appear to be a clustered addition to an emerging
necropolis of the eleventh century and onward, expanding upon the vast
graveyard of Chira that stretched 2 km to the south.79 Here, at least
two single-room, originally domed, tomb-like structures contrast to the
smaller graves blending into once cultivated fields, indicating that this
landscape had begun to be memorialized since the eleventh century,
extending northward with the addition of the four Lal Mara tombs.
Beyond stylistic considerations, none of these structures – either at Chira
or at Lal Mara – have inscriptional or other elements that can help in
dating them more definitively.
However, the extensive presence of blue and white glazed tiles in the
decoration of all four of the Lal Mara tombs has been the focal point of

chronological considerations. There is scholarly consensus that the greatly

varied and experimental nature of the molded and incised motifs on the
tiles, and the unevenness of their glazing – resulting in gradations espe-
cially in hues of blue – are hints that the Lal Mara tombs were among
the earliest instances of buildings with glazed-tile decoration along the
Indus’s alluvial areas.80 Although it is more certain that this decorative
technology reached its apogee in the monumental tombs erected in the
city of Multan throughout the fourteenth century, when precisely this
technology was introduced – and consequently the date of the Lal Mara
tombs – still remains a debated question. Some scholars have argued for
an eleventh-century introduction of the tile technology, citing the numer-
ous visitations to the region by the Ghaznavid sultans, either directly to
Indus cities such as Multan or en route to India, during which construc-
tion and tile application could have taken place. Others have proposed
a mid-thirteenth-century date instead, as “the transfer of this craft from
Afghanistan and Khorasan likely occurred in the wake of the Mongol dep-
redations,” which brought many skilled craftspeople into Indus-contiguous
regions as they sought new patronage.81
There is the third, intermediate possibility of dating the Lal Mara tombs
to the last two decades of the twelfth century, as the Ghazna Shansabanis
were consolidating the entire region. This possibility is raised by the
recent scholarly re-dating of the Shansabanis’ key architectural activities in
Afghanistan, namely, the magnificent minar towering over the Firuzkuh
Shansabanis’ sard-sir encampment or summer “capital,” and the Ghazna
Shansabanis’ renovations and modifications at Ghazna and Lashkari
Bazar (cf. Chapters 3–5). Additionally, our own examination earlier in
this chapter of the Ghazna Shansabanis’ other architectural traces in the
middle Indus area, and at Sukkur in Sindh, lends contextual support to
this proposed re-dating of the Lal Mara tombs.
The origins of a propensity toward commemorative structures among
the Ghazna Shansabanis and their kin-based and other dependents was
discussed above, as a possible continuation of tendencies embedded in
what was actually a quite recent, more transhumant and/or pastoralist
history. Given the probable non-royal but still elite status of the patrons,
single-room tombs rather than more imposing mausolea were the pre-
ferred scale of construction. Moreover, it was equally discernible that,
rather than importing pre-designed architectural ideas for their needs,
these patrons had relied on locally based building traditions, cumulatively
creating an unprecedented variety among structures dating to the same
two decades. This primary reliance on localized architectural conventions
was consistent with an overall modus operandi of adaptive consolidation
and re-inscription of their new territories, rather than the imposition of
imperial command on them. This perspective, then, allows for the inclu-
sion of the Lal Mara tombs as part of the Ghazna Shansabanis’ activities

322 Iran to India
Figure 6.38 Lal Mara Sharif, Tomb IV: exterior, west façade, detail of glazed-tile
decoration. © Photograph Alka Patel 1997.
in the region, fitting within the architectural diversity they had already
demonstrated in their patronage.
The introduction of glazed-tile technology into the area of the north-
ern Indus’s alluvial plains is, admittedly, difficult to date with precision.
Preliminary studies of the glazed tiles recovered at Ghazna by Italian archae-
ologists in the 1950s concluded that the production of this decorative mate-
rial likely began during the eleventh-century highpoint of the Ghaznavid
empire, ceasing temporarily as of c. 1150 ce and ‘Ala’ al-Din Husain
Jahan-suz’s destruction of the capital. Shortly thereafter, Ghazna was occu-
pied by bands of Oghüz nomads, whose peripatetic presence and irregular
use of the site did not call for glazed-tile production. But since the tiles were
found in the upper and thus later occupational strata of the Ghazna palace,
Scerrato proposed that the Shansabanis re-initiated glazed-tile-making
there after their definitive occupation of the city in the mid-1170s (cf. also
Chapter 5).82 Furthermore, the re-dating of the Jam-Firuzkuh minar to ah
568/1174–1175 ce (previously ah 588/1193 ce) – likely serving as a commem-
oration of the Shansabanis’ victory at Ghazna – also has consequences for
our understanding of the dissemination of glazed-tile usage: the presence
of bright blue tiles on the minar, emphasizing the inscriptional band of the
Firuzkuh sultan Ghiyath al-Din’s royal titles, indicates that the technol-
ogy had made its way eastward from Saljuq Iran (cf. Figure 1.9) at least a
half-century prior to the Mongol campaigns of the 1220s. The conveyance
of glazed-tile technology to the upper Indus’s shores, then, could have
occurred sometime during the eleventh through twelfth centuries.

A final, important point for consideration is the decorative iconogra-

phy of the glazed tiles on the four Lal Mara structures, and its distinctive-
ness from the decorations of the other Shansabani-period structures along
the Indus. Akin to the Sukkur tombs (see above), at Lal Mara also there
is no epigraphic program. But in contrast to the former, the Lal Mara
tombs carry no pseudo-epigraphy, either.83 Instead, the glazed blue tiles,
forming horizontal bands on the tombs’ exterior façades, carry varieties
of rustically rendered geometric and curvilinear motifs (Figure 6.38). The
tiles were complemented by glazed and unglazed ceramic plugs in a range
of geometric shapes, framing doorways and appearing extensively on the
interior surfaces of the tombs; but the tiles themselves appear to be uni-
formly square.
These characteristics differ greatly from the glazed tiles recovered from
the central palace at Ghazna, which came in a plethora of shapes and were
decorated with a wide variety of floral, arabesque, and zoomorphic motifs.
Also, the Ghazna decorative iconographies were rendered in a uniformly
polished manner, and could be termed cosmopolitan in that they formally
echoed visual–organizational practices that were transregionally recogniz-
able across the Persianate regions (cf. Chapter 5). Such cosmopolitanism,
then, formed a stark contrast to the localized, or “vernacular,” shapes on
the Lal Mara tiles, which hearkened to very different iconographic tradi-
tions and stylistic practices.
R. Azeem had already noticed the uniqueness of the Lal Mara tiles
when he suggested that their inspirations lay outside the decorative tradi-
tions documented along the Indus’s alluvia, and certainly beyond the more
cosmopolitan traces surviving in Khurasan during the eleventh through
thirteenth centuries. He further observed that “there was clear evidence of
[the Lal Mara tombs’] strong relationship with the pre Muslim [sic] tombs
of … Baluchistan.” But Azeem did not pursue the concrete possibilities of
this smaller-scale but still transregional connection, claiming the absence
of direct routes of communication between Makran–Baluchistan and the
areas around the upper Indus. 84
The Baluchistan link must be revisited for both historical–geographical
and formal reasons. The cultural and physical geographies of Baluchistan
and the upper Indus, and routes of travel between them, may in fact be
more difficult to traverse now, than they were a millennium ago. The
centers of Kalat, Kharan and Panjgur in Baluchistan, and even the Makran
coast were all communicable along the entire course of the Indus.85
Moreover, as we saw above, the influx and expansion of the Saljuqs and
their dependent clans into the Iranian Plateau during the tenth through
twelfth centuries instigated in turn the large-scale migrations of Baluch
tribal confederacies from Kirman eastward to the Makran coast and its
hinterland, ultimately giving their name to the region. But many – if not
the majority – of Baluch tribes continued to be nomadic pastoralists, pur-

324 Iran to India
suing further migratory pathways connecting Makran and the eponymous

region of Baluchistan with inland areas in seasonally and commercially
driven transhumances.86
Moreover, the typological, iconographic, and stylistic resonances
between the Lal Mara and Baluchistan tombs should also be re-examined.
As discussed earlier in this chapter, the pre- and early Islamic tombs in
Baluchistan consisted of the familiar domed cube with no outwardly visible
transition zone – thus creating formal parallels between, for example, the
Lasbela tomb attributed to Muhammad ibn Harun (Figure 6.10) and Lal
Mara’s Tomb III (Figure 6.39). Also mentioned above were the unidenti-
fied, possibly earlier terracotta-clad tombs near Baluchistan’s settlements
and towns, such as Kalat, and scattered in less trafficked landscapes: these
domed-cubed burial structures also had no exterior transition zone and
were virtually covered with square, unglazed terracotta plaques, placed in
horizontal registers separated by decorative friezes. Indeed, the Lal Mara
tombs’ patterned placement of blue glazed tiles over much of their exte-
rior surfaces echoes the unidentified Baluch tombs more closely than the
“textile-like bands” of the Lasbela tomb attributed to Muhammad ibn
Harun. This juxtaposition of the surviving structures’ variations on the
domed-cube architectural typology, and especially the surface decoration
among all of them, only reinforces the Baluchistan–Panjab “relationship”:
the Lal Mara tombs both maintained the resonances with their Baluch
parallels (Tombs I and II [Figure 6.40]), while also complicating the
plans to hide the dome-cube contact behind octagonal facing (Tomb IV
[Figures 6.38, 6.41]).
In the end, surely it would be unsurprising that the Shansabani-affiliated
military leaders campaigning along the upper Indus would commission
funerary structures adhering to the basic domed-cube typology with the
rustic decoration described above. The active demographic connections
between Baluchistan and western Panjab, maintained by Baluch and
perhaps other nomadic pastoralists via their seasonal migrations, would
have communicated the localized, “vernacular” aesthetics of such rustic
motifs and styles across the two regions. Moreover, the incipient technology
of blue-glazing was evidently available and favored along the upper Indus.
There is no reason that the Shansabanis’ by now well-evidenced tendency
to patronize the prevalent building traditions within an architectural–
cultural region – rather than importing traditions from without – would
not have been followed here. The Shansabani patrons of these funer-
ary structures called upon the locally available architectural and deco-
rative practices for commemoration as they campaigned in the western
Panjab.

Figure 6.39 Lal Mara Sharif, Tomb III: exterior, west façade. © Photograph Alka Patel 1997.
Spanning Imperial and Architectural Spectra
This chapter’s examination of the Ghazna Shansabanis’ architectural

traces in the Indus-dependent areas affords unique insights regarding their
methods of building and consolidating their imperial presence throughout
the region. These insights will be key considerations as we accompany them
farther eastward on their forays into the north Indian duab in Volume
II. Indeed, the Indus-dependent areas remained seminal well beyond
the period of Shansabani presence there, serving as connective pathways
between the eastern Iranian and western Indic cultural worlds, particularly
for subsequent seekers of empire in the Indian subcontinent, during nearly
a half-millennium into the sixteenth century (cf. Introduction).
The marshalling of evidence in this chapter has allowed us to observe
that re-inscription continued to be Mu‘izz al-Din’s preferred modus of
establishing his imperial writ in these newly occupied areas. Chapter
5 argued that the Shansabanis re-inscribed the imperial centers of the
Yamini–Ghaznavids throughout Zabulistan and Zamindawar. Vast archi-
tectural complexes, with still salvageable architectural fabrics, were reno-
vated and perhaps even reoccupied as residences. These evidentiary data
signaled not only the Shansabanis’ regional ascendancy, but also provide
us with hints at shifts from tent to palace among some of the elites.
In the plains and valleys contiguous to and dependent upon the

326 Iran to India
Figure 6.40 Lal Mara Sharif, Tomb I, exterior, west façade. © Photograph
Alka Patel 1997.
Figure 6.41 Lal Mara Sharif, Tomb IV, exterior, west façade. © Photograph Alka Patel 1997.

Indus, however, Mu‘izz al-Din and his military commanders imple-

mented re-inscription on a scale unseen in the Yamini–Ghaznavids’ erst-
while imperial centers. Here, the Shansabanis encompassed entire urban
areas and their hinterlands by means of re-inscription, so that important
regional centers such as Multan and Lahore came under the Shansabani
aegis without extensive new architectural projects. These cities and their
surrounding landscapes had received a Perso-Islamic imperial “infrastruc-
ture” during the preceding nearly two centuries, as the Ghaznavids had
counted the territories through the western Panjab as part of their empire
until their final defeat at the hands of the Shansabanis in ah 581/1185
ce. Mu‘izz al-Din astutely perceived the speed, economy, and efficacy
of re-inscribing existing buildings, institutions, and cities by means of
his own endowments, rather than having to build spaces and populate
them anew (cf. above in this chapter). While re-inscription furnished
enormous advantages for the Ghazna Shansabanis, it problematizes the
modern investigator’s task: Mu‘izz al-Din’s interventions, even on a scale
aggrandized beyond architectural complex to urbs-hinterland, are that
much more difficult to glean and thus understand.
Nevertheless, this chapter analyzed the structures that are epigraphically
attributed to Shansabani presence in the Indus region, or can be associated
with it otherwise. As already noted, none of the surviving Indus buildings
apparently resulted from royal patronage – though clearly ensuing from
patrons with the means to access skilled laborers. The riba† outside Multan
might come closest to royalty, given its commissioning by the Mu‘izzi
sipahsalar ‘Ali ibn Karmakh. The other two tombs in Multan’s vicinity
were also quite possibly patronized by nobility (cf. below). Although not
conclusive, it is reasonable to propose that, in the case of Multan, the sights
of royal patronage had been focused on re-inscription of the city’s major
architectural complexes as Shansabani institutions. Thus, the hinterland
was relegated to Mu‘izz al-Din’s nobles to make their own marks – though
collectively, these would still accrue to an overall Shansabani authority in
the region.
The Indus structures’ respective locations and discernible features
united them into three distinguishable groups. The Multan-area tombs
included decorative elements traceable to the Gandhara–Nagara archi-
tectural culture, combined with prominent epigraphic programs. The
latter frequently included Qur’anic verses and doxologies known from
Shansabani buildings farther west in their home regions, and among their
conquests in Zabulistan and Zamindawar (Chapters 4 and 5). Meanwhile,
Kufic-style pseudo-epigraphy graced the Sukkur buildings, along with
other decorative elements that placed them within the Maru–Gurjara
ambit. Finally, in overall form and decorative practices, the Lal Mara
tombs evinced connections with funerary architecture in the contiguous,
southwestern region of Baluchistan. Paradoxically, the immense variety

328 Iran to India
in these surviving structures’ modes of manufacture attests to what might

be termed the Shansabanis’ architectural modus operandi: the three groups
each adhered to their respective architectural–cultural environments,
demonstrating the Shansabani-affiliated patrons’ consistent reliance on
localized skilled laborers. Furthermore, the remarkable variety among
these structures underscores the variegation in style and materials imbuing
the western frontiers of the Indic sphere, providing a salutary caution
against circumscribing this region within any artificial homogeneity.
Based on these discernible features, it is tempting to cleave the
Shansabani-attributed architecture throughout the Indus regions into iter-
ations of “cosmopolitan” and “vernacular” building traditions. Certainly,
the Multan-vicinity structures seem to espouse a recognizable (if incipient)
imperial aesthetic, thereby fitting within the “cosmopolitan” category,
just as the Sukkur and perhaps especially the Lal Mara tombs deviate
from it. However, the above caution regarding artificial homogeneity is
equally relevant here: the exigencies of varying cultural geographies in this
overall region provided a plethora of building traditions that were robustly
area-specific, mitigating against sudden impositions of new architectural
ideas. This reality must be placed alongside the imperial culture of the
Ghazna Shansabanis – wherein, for example, re-inscription superseded
imposition. Rather than being monolithic and rigid, such an imperialism
evinced a notion of authority that was contingent and spontaneous. These
two factors together contributed to the continued variety in built forms,
even as the Shansabanis crafted an empire that was ultimately more ambi-
tious and encompassing than that of their renowned Yamini predecessors.
Notes
1. ‘Ali ibn Karmakh was probably appointed wali of Multan sometime after the
Shansabani campaign to the Multan region in ah 571/1175–1176 ce (see also
main text). However, the sipahsalar was appointed to Lahore as of ah 582/1186
ce, the year of the third and final Shansabani campaign in the Panjab, which
definitively signaled the end of the Ghaznavid house. Cf. Juzjani vol. I, 1963:
398 (trans. vol. I, 1881: 456).
2. Taj Ali’s doctoral dissertation (Bonn, 1987) is the only known, focused
and thorough treatment of the site, later published as a monograph in the
series Memoirs of the Department of Archaeology, University of Peshawar
(Ali 1988). See also Edwards 1990: 231–254, 412–428; Edwards 2006: 27–28;
Edwards 2015: 145ff., 211–229 (in the last work, the author’s reconsideration of
the site resulted in her attributing it firmly to the thirteenth century). A brief
description of the four Lal Mara tombs appears also in Dani 2000: 560.
3. See A. N. Khan 1988: 306–307, 311–316, 322; M. Kevran 1996b: 136–143. Upon
comparing the Sukkur tombs with the so-called tomb of Muhammad ibn

Harun near Lasbela (ancient Armabil) in Baluchistan, Khan dated the Lasbela
structure earlier, the original possibly erected upon the death of Muhammad
ibn Harun, who had been in the entourage of Muhammad ibn Qasim during
the early eighth-century ‘Umayyad entry into Sindh. However, he attributed
the cut-brick decoration on the structure’s exterior to a later date when the
tomb was repurposed, viz. “used for [a] second burial which might belong
to the Ghaznavid period of the eleventh century.” Kevran expanded the date
span of these structures to the tenth–twelfth centuries, based on comparisons
with surviving Central Asian tomb structures and their decorations, which
were nonetheless supplemented by “certain exotic characteristics that lead
one to think that Central Asian influence was not alone in determining their
ornamentation.”
4. Bosworth 1966: 87.
5. For the Ghazna palace, see esp. Allegranzi 2019a, vol. I: 146ff.; also Juzjani
vol. I, 1963: 362 (trans. vol. I, 1881: 384). Cf. also Chapters 4 and 5; Appendix,
Afghanistan V.
6. Ziad 2016; Anooshahr 2018b: 12.
7. Cf., e.g., A. N. Khan 2003: 14; Anooshahr 2006: 278, 286ff.; and esp.
Anooshahr 2009: 62–73, where the precedents for ghazw and Mahmud’s own
role in utilizing them in constructing his image for posterity are explored.
Ziad (2016: 651) also questioned “the political motives behind iconoclastic
narratives in primary and secondary sources.” Notably, Anooshahr (2018b:
26–27) has proposed that Mahmud’s use of war elephants, particularly in
westward Ghaznavid campaigns into Khurasan and beyond, created an
identity problem of sorts for the Ghaznavid rulers: the elephant’s diaboli-
cal associations in the Islamic ecumene, paired with its formidable and even
undefeatable value as a “weapon,” presented a dilemma that could be resolved
only with emphatic ghazi credentials, viz. the destruction of temples, or idol
worship in its most tangible form. Thus, “the reports of temple desecrations
in India enter the various accounts [of Ghaznavid campaigns] only after ele-
phants appear on the scene” (original emphasis).
8. Cf. esp. Patel [2004] 2007 for an overview of the pivotal role the coast
of Gujarat played in the Indian Ocean world during the eleventh through
fifteenth centuries; also Thapar 2004: 18–37. For a reinterpretation of the
Ghaznavids’ Somanatha campaign as one of longer-term “conquest” rather
than plunder, cf. Patel 2005. For a parallel re-examination of the Ghaznavids’
forays into Gandhara-Swat as more than temple raids, viz. as “pursu[ing]
long-term objectives”, see Ziad 2016: 652 and passim; and infra in main text.
9. For the India campaigns of Mahmud and Mas‘ud I, see Bosworth 1973: 75–76,
235; Inaba 2013: 78–79. For the re-initiation of campaigns into the north
Indian duab during the reign of Mas‘ud III (1099–1112), see esp. Bosworth
[1977] 1992: 85–87; and Chapter 4.
10. E.g., Mahmud Ghaznavi’s firm control of the Gandhara–Swat region by
means of mosques, madrasas, and rest-houses; the site of Manikyala, e.g.,
was described by Gardizi ([trans. Bosworth] 2011: 111–112) as in continual use
through the reign of Mas‘ud I. See also Siddiqui 2010: 11–12.
11. A case in point would be of course the aforementioned ‘Ali ibn Karmakh,

330 Iran to India
deputed to Multan and its vicinity, and eventually Lahore. Cf. infra in main
text.
12. A century earlier during the late eleventh century, Lahore was already
known as “little Ghazna,” due to its political and cultural significance for
the Ghaznavids: the important Ghaznavid poet Mas‘ud Sa‘d Salman (c.
1046/1049–1121 ce) was not only born there, he was “a courtier and a poet
and an established member of the early Iranian aristocracy base in India”
(Sharma 2000: 15–16, 19–20). For continued Ghaznavid focus on Lahore, see
also Allegranzi 2019a, vol. I: 43–46.
13. Cf. esp. G. M. Khan 1991: 129; A. N.2003: 14, 17; Flood 2002: 104–105;
Jackson and Andrews 2012. For greater detail regarding the transformation of
Lahore from a garrison town to a provincial Ghaznavid capital of culture, see
Edwards 2015: 41–43, 46–52. See also Chapter 4, above.
14. Bosworth (1973: 75–77) described Ghaznavid attempts at a “dual administra-
tion” as early as the closing years of Mahmud’s reign, more closely linking
at least the Panjab with the central Ghaznavid lands. The sultan posted
Ghaznavid military and legal authorities there to administer the area directly,
and the arrangement seems to have endured through some part of the reign
of his son Mas‘ud I. Meanwhile, the north Indian duab was too distant from
Ghazna to be administered as an annexed territory, and tribute from the
region to the Ghazna treasury came principally in the form of plunder from
the near annual campaigns, which were prevented in some instances by means
of a hefty advance payment, as had been the case with the city of Multan
during Mahmud’s campaign there in ah 395/1005–1006 ce (cf. also infra in
main text).
15. Cf. esp. Deloche (trans. Walker) vol. I, 1993: 24–27; Edwards 2006: 22; Asif
2016: 41, 49–50; and Chapter 2.
16. See esp. A. N. Khan 1983: 177; Friedmann and Andrews 2012; Edwards 2015:
28–31. However, al-Biruni recorded that Halam ibn Shaiban had replaced
the ‘Umayyad mosque with his own residence; the mosque was rebuilt by
Sultan Mahmud Ghaznavi after his entry into the city in ah 395/1005–1006
ce (quoted in Asif 2016: 112).
17. Isma‘ili missionary activity in Sindh began as early as the later ninth century,
likely even before the establishment of the Fatimid capital at Mahdiyya in
Tunisia in ah 297/909 ce. Cf. Flood 2009: 50ff.; Friedmann and Andrews
2012; H. A. Khan 2016: 7–8; Asif 2016: 45.
18. Cf. A. N. Khan 1983: 177, 317.
19. Edwards (2015: 103) summarized the proposed re-inscription of inherited
landscapes and built assets by the Shansabanis as “a certain courtly contin-
uum.” Cf. also Chapters 2, 3, and 5.
20. While Deloche (vol. I, 1993: esp. 26 [trans. Walker]) marshalled historical evi-
dence toward delineating major routes, recent work by Neelis (e.g., 2011: 230
and passim) has explored complementary sources such as historical “graffiti”
to trace the seasonal and shifting capillary routes in the mountains beyond the
upper Indus.
21. See esp. Rahman 1979: 33, 52, 228 and passim. Even during the rise of the
Hindu-shahi rulers in the mid-ninth century, Arab-Muslim contingents

– splintered from the southern ‘Umayyad–‘Abbasid forces in Sistan and its

environs – were resident in the northwest, as evidenced by at least four
known bilingual inscriptions from Swat – cf. notes infra; also Chapters 2
and 3.
22. The Italian excavations of the site have done much to clarify its chronol-
ogy and some cultural activity. The mosque itself underwent two phases of
construction, the latter likely in the mid-eleventh century, according to an
Arabic inscription on a reused decorated marble slab. The cemetery west of
the mosque – behind the mihrab in fact – unearthed a conspicuous shrine of
a revered personage, and an overall sequence spanning the twelfth and thir-
teenth centuries; curiously, some graves contained both human and animal
bones – an unusual and unorthodox Islamic practice, “probably the result of
… some local beliefs” (Bagnera 2006: 225). See also Rahman 1979: 275–276;
G. M. Khan 1985; Manna 2006; Giunta 2006.
23. Scant remains of Ghaznavid-period intervention at Giri, near Taxila (ancient
Takshashila) consisted of an enclosure with rooms along the perimeters –
possibly a caravanserai or riba† – and a small mosque in the large central
courtyard-like space. It is possible the mosque alone was of Ghaznavid con-
struction, as the remaining structure’s extremely practical layout was a staple
among the region’s civic–public complexes since at least the mid-first millen-
nium ce. Interestingly, the complex’s architectural fabric consisted of kanjur
and lime mortar, the distinct local building materials that were standard in
the Gandhara–Nagara ambit. Cf. Rehman 1991: 38–40; A. N. Khan 2003:
14–15, 17; and see infra in main text.
24. E.g., Meister 2010.
25. Cf. Meister 2010: esp. 12–13.
26. See esp. Meister 1996; Meister 1997/98: 45ff.; Meister et al. 2000; Meister
2010: 58–59.
27. See also Meister 1997/98: 45; Meister et al. 2000: 41.
28. Meister 2010: 33.
29. Earlier skepticism regarding the identification of Banbhore as Dibul, the
eminent Indus port city, ensued in part from the seemingly late foundation
of the large “Arab-style” congregational mosque at the site (Figure 6.9), in
which the earliest of the thirteen plaited Kufic inscriptions bore the date of ah
109/727 ce (cf. F. A. Khan 1963: 16ff; A. N. Khan 2003: 2–4). For an overview
of the textual sources (ninth through eighteenth centuries) regarding histori-
cal Dibul, see esp. Kevran 1999b: 81–89, who clarifies that the modern debate
surrounding the identification of Banbhore as Dibul has been stoked by the
oft-shifting course of the Indus itself; see also Patel 2011. Indeed, the identifi-
cation of Banbhore–Dibul is supported by surrounding sites, e.g., Ratto Kot,
preceding Banbhore–Dibul on the west and serving as “a first line of defense
and communication for the rich commercial centre …”: Kevran 1993: 17 and
passim; cf. also Kevran 1992: 148–172.
30. See esp. Patel [2004] 2007: 36–40; Flood 2009: 15ff. See also Asif 2016:
115–116.
31. Cf. Kevran 1993: 26–37 for the accumulation of forts and “fortified warehouses”
that served various functions over time in relation to Dibul, particularly as

332 Iran to India
the Indus’s course shifted farther east and created new configurations of its
complex débouchement into the sea; cf. also Kervan 1999a: 150–152; Kervan
1999b: 116–118 for a proposed sequence of the gradual displacement of Dibul
as the pre-eminent Indus river port city, and its dwindling importance and
de-emphasis in historical sources through the seventeenth century.
32. Much important work has been done on the pre-Islamic networks linking
the western through eastern Indian Ocean littorals (see Bibliography);
works such as Daryaee 2003 and Bopearachchi 2006 provide concentrated
foci on maritime communication between West and South Asia; see also
Patel 2011 for an overview of the available evidence and its implications.
As recently reiterated by Asif (2016: 26ff.), the very vastness of the “Indian
Ocean world” – from the Mediterranean through Southeast Asia – impedes
any treatment encompassing all of its shores and dependent societies across
the centuries, not least since smaller and larger networks and their nodes of
contact shifted over time. Thus, even though the entirety of the subcontinent’s
coastlines were imbricated within the transoceanic routes of the Indian Ocean
world, it is most fruitful to conceive of and concentrate on tighter networks,
e.g., Sindh “as an Indian Ocean region … long connected with Arabia …
contain[ing] settlements, trading connections, and ports that predated the
birth of Islam … and continued after the rise of Muslim political power in
the region” (ibid., 32, 44–46). It was observed supra in the main text that
the oceanic interconnections did not operate in isolation on a macro level,
being in essence the maritime complements to intricate overland and river-
ine networks: a granular view of these regional connections can be gleaned
from a variety of cultural artifacts, including poetic works such as the late
twelfth-century Samdesharasaka, composed in Prakrit probably at Jaisalmer
(modern Rajasthan, India), which “reveals mercantile geography along the
path anchored by Uch and the long string of Cholistani forts [to Khambhat,
historically on the Gujarat coast]” (ibid., 71–73).
33. Cf. Patel [2004] 2007: 55; Asif 2016: 49–50.
34. See esp. Pehrson 1966: 2; Baloch 1991: 247–248; Spooner [1988] 2010: pts 3
and 4; Edwards 2015: 180–181.
35. Muhammad ibn Harun, the personage putatively commemorated here, was
a military leader of the ‘Umayyad forces in Sindh–Makran during the early
eighth century; but it is improbable that the structure was contemporaneous
with this historical figure, since monumental funerary commemoration was
not characteristic during the Umayyad caliphate, nor would an active military
frontier have been a propitious site for such a building project. See esp. A. N.
Khan 1988: 305–307; A. N. Khan 2003: 38–39; Dani 2000: 561; Edwards 1990:
esp. 366–371; Edwards 2006: 21–25; Edwards 2015: 175–183.
36. See esp. Hassan 1991: 78–85, and figs. 2–9 – among the only published images
of these fascinating mortuary structures. Without argumentation, A. N. Khan
(2003: 50–51) has considered other domed structures in Baluchistan, abun-
dantly clad with decorated terracotta plaques but eschewing figuration, as
datable to the fourteenth century as the tombs of the so-called Nikodari
– perhaps referring to the trailing Mongol contingents of Negüdaris (cf.
Chapter 4 notes). These tombs’ decorative plaques contrast with above-

mentioned, smaller ones in the Panjgur vicinity (northwestern Baluchistan)

carrying simplified equestrian and other scenes with rudimentary figures.
37. For the Baluch groups and their occupations, see esp. Pehrson 1966: 4, 10–11,
16; also Amiri 2020: 157 for the Helmand Baluch, within whose ambit above-
ground vaulted rooms for entombment were also documented. Cf. Chapter
2, for a comparable picture of the pre-imperial Shansabanis and other
Ghuris.
38. See Edwards 1990: 366–369; Edwards 2015: 178–179.
39. Explorations of Tor-Dherai, northeastern Baluchistan, revealed more loose
terracotta plaques with pre-Islamic Buddhist–Brahmanical motifs such as
swastikas, and plastically rendered floral elements also evoking stupa or temple
iconographies; these are comparable with the terracotta plaques from the
great stupa of Mirpur-Khas (eastern Sindh). Apart from the Tor-Dherai and
Mirpur-Khas finds, it is difficult to posit even a relative dating for the remain-
ing plaques. See Cousens 1929: 82–97 and esp. pl. XXIV; Stein and Konow
1929: figs. 25–28; Stein and Konow 1931: 38 and fig. 5.
40. See esp. Kevran 1993; Kevran 1999b.
41. Cf. Ibrahim and Lashari 1993: 19–24; Kevran 1993: 22–24; Kevran 1996a:
46–47; Kevran 1999a: 148–150; Kevran 1999b: 72–73.
42. Cf. Cousens 1929: 126. For eleventh- and twelfth-century temple remains in
the vicinity of Bhodesar (Tharparkar district, Sindh), see also Raikes 1859:
10–12, 83–84. I am grateful to Dr. Fatima Qureshi for providing photographs
of Tharparkar’s architectural remains; her monograph on the architecture at
the vast Makli necropolis and its sources is eagerly awaited.
43. Flood 2009: 46.
44. See esp. Dhaky 1968: 68, 71ff.; Patel [2004] 2007: 79, 93; Patel 2015: 88–90.
The Maru–Gurjara style (cf. infra in main text) is still most often associated
with temple architecture; its dissemination within India, alongside the move-
ment of Jaina communities during the twelfth through sixteenth centuries
and into the present day, has most recently been studied by Hegewald (2015).
45. M. A. Dhaky (1967) first used the term “Maru–Gurjara” in discussing the
magnificent temples at Kiradu (Barmer district, southwestern Rajasthan cf.
Volume II). In a later work (Dhaky 1975), he charted the combination of
temple-building practices in central–southern Rajasthan (ancient Maru-desha),
and mainland Gujarat and adjacent Saurashtra (ancient Gurjara-desha) during
the later tenth–eleventh centuries to form what he called the Maru–Gurjara
style. It should be noted that, consistent with the emphasis on visual analysis
and the identification of distinct styles that were prevalent in 1960s–1970s
art-historical scholarship – particularly on historical South Asia – very little
investigative energy was spent in identifying the historical mechanisms for
such stylistic developments (e.g., changes in taxation or other factors leading
to migrations of artisan groups etc.). Such work has been undertaken through
a study of inscriptions surviving from earlier periods in the region (cf., e.g.,
Ray [2004] 2007), demonstrating a possible approach for later centuries as
well. Despite scholarly attention being focused on the Maru–Gurjara canon
and its treatises, the work of A. Hardy (e.g., 2015) has initiated what will
hopefully develop into equally detailed scholarship on the region of Malwa in

334 Iran to India
north–central India, where the architectural treatise Samaranganasutradhara

was composed during the eleventh-century reign of the legendary Paramara
ruler Bhojadeva (r. c. 1010–1055). Cf. also infra in notes.
46. See Patel [2004] 2007: 5ff.
47. Although beyond the purview of the current book focused on the Shansabanis
and their “slice of time” – which they admittedly superseded in terms of their
brief ascendancy’s long-term impact (cf. Introduction, and esp. Volume II)
– it is worthwhile noting here that a distinct and abundant body of stone
architectural remains, surviving from Sindh to western Makran–Baluchistan,
urgently call for analysis. Dani (2000: 571) referred to the remnants at Makli
as “the Thatta school of architecture in Sind.” But I would argue that, collec-
tively, these remains provide a unique and fascinating opportunity to trace
the remarkable innovations within the Maru–Gurjara style (cf. infra in main
text), as its practitioners applied it to the ritual and architectural developments
within South Asian Islam. While temples and mosques ensuing from the
Maru–Gurjara tradition still stand at several sites in Sindh’s Tharparkar dis-
trict, extensive funerary remains, ranging from imposing mausolea to small-
scale stone sarcophagi and cenotaphs, at sites such as Makli (near Thatta),
Chaukhandi, and continuing farther west at Taung, Jerruk, and eventually
Lasbela (Baluchistan), are arguably significant developments of the building
practices prescribed by the Maru–Gurjara canon. For Kalat and its environs
in central Baluchistan, see the albeit brief descriptions of Charles Masson
1843: 389–390. Documentation and initial analysis of many related remains
in Sindh have been conducted by Zajadacz-Hastenrath 1978; Baloch 1991;
Bokhari 1992. Indeed, all of these gradually differentiating yet somehow relat-
able architectural developments beg the question of the validity of a singular
“style,” and the concept’s ability to encompass socio-religious communities,
regions, and centuries.
48. For some extremely important but neglected stone remains of chhatri tombs
at Gwadar (western Makran), see Stein and Konow 1931: 72–73 and figs. 19
and 20 (Masson for central Baluchistan, mentioned supra in notes). These
remnants’ evidently distinct stone treatment, both as a building material and
medium of decorative expression, calls for comparison with similar tombs at,
e.g., Bhadreshvar and Makli, and strongly suggest that their construction was
carried out by Maru–Gurjara trained artisans. These remains would mark the
farthest western reach of the Maru–Gurjara canon. Cf. also elsewhere in notes
for coastal and overland connections linking Gujarat, Sindh, and ultimately
the continuing coastline westward. Certainly, the Baluch nomads would have
encountered and contributed to the juxtaposition of clearly distinguishable
vernacular and elite architectural activity within the same landscape (cf. supra
in main text and notes) – a tendency quite possibly continued by the incom-
ing Shansabanis.
49. See esp. Dhaky 1968: 67–68, 72ff.; Patel [2004] 2007: 83ff.; Patel 2015: 89.
The contiguity of the regions of Rajasthan–Gujarat and Malwa may have
been attested in their respective treatises: Dhaky (1968: 68) noted that another
manuscript titled Jayaprrcha, surviving in fragmentary form, had been pro-
duced in Malwa, also in the twelfth century and perhaps earlier. While noting

this work’s origins, Dhaky did not, however, address the question of whether
the text actually contained the same content as its Maru–Gurjara namesake,
or simply bore the same name even though its content differed (not uncom-
mon in pre-modern textual practices). This would be an important point,
since if the former was the case, then like the Maru–Gurjara Jayaprrcha, the
Malwa Jayaprrcha would have also contained prescriptions for constructing
mosques, albeit conforming to the conventions localized within its regional
purview.
50. See esp. Patel [2004] 2007: 79ff.; Patel 2015: 88; Desai 2012: 472ff.
51. Kevran 1999a: 149.
52. The inscriptional fragments from the site consist of a small fragment in situ
to the left of the mihrab of Thamban Wari masjid and other loose fragments
retrieved from the larger congregational mosque. However, they provide too
little information regarding calligraphic style or content to warrant secure
dating: although Ibrahim and Lashari (1993: 15–21) compared the fragments
with Ghaznavid and later architectural epigraphy in Afghanistan, Allegranzi
(pers. comm., April 2019) found them comparable to earlier architectural
inscriptions of the eleventh century.
53. For the early development of Sufism, of course the “classic” reference has
tended to be Trimingham 1971: esp. 5–18; see also Karamustafa 1994: 3–5
for the more antinomian of the renunciant orders. H. A. Khan (2016) has
concentrated on the Indian developments of Sufi orders and their until now
little-known connections with Shi‘ism and particularly Isma‘ilism.
54. A prime example of devotional longevity is the tomb of Sheikh Yusuf Gardizi
(d. ah 546/1152 ce) (Figure 6.15), whose shrine was tile-clad and much
expanded by the sixteenth century and has continued to have hereditary
caretakers into the twentieth century (see Mumtaz 1985: 43–44; Edwards
2015). The sheikh arrived in Multan about seven decades after Mahmud
Ghaznavi’s persecution of the city’s Isma‘ilis and its collateral destruction.
His arrival seems to have coincided with and fomented the revival of Multan,
providing “renewed vigor around another non-Sunni [i.e., Sufi] focus.” See
esp. Edwards 1990: 74–75, 246.
55. Such was the case with another one of Multan’s early Sufi sheikhs, Baha’ al-Din
Zakariya (ah 578–661/1182/3–1262 ce), the founder of the Suhrawardiyya in
India whose khanaqah in the city was “a place of great political and strate-
gic significance” (Nizami [1961] 1978: 221ff., 240). Thus, the stark contrast
between the Chishtiyya and the Suhrawardiyya regarding contact with figures
of political authority was established from the entry of the latter into the
region. See also Sobieroj 2012; F. A. Khan 2016: 31–35.
56. Cf. Hüttel and Erdenebat’s (2010: 4–5) reflections on past archaeological
studies of the Uighurs at Karabalgasun, as they reported on their own find-
ings. See also Azeem (1991: 89–90), who touched upon the small early tombs
(tenth–eleventh century?) in Baluchistan, attributing these to Muslim Baluch
tribes whose nomadism “was not likely to have developed … monumental
buildings.” Cf. also infra in main text and notes.
57. See esp. Edwards 1991: 89ff.; Kevran 1996: 154ff.; Flood 2002: 132ff.
58. For cogent scholarship on floral and zoomorphic iconographies and

336 Iran to India
symbolism, see, e.g., Falk 2006: 145–146; also Dallapiccola 2018; Smith 2018.
These studies update but also still complement the vast data contained in
earlier works, such as those of James Fergusson (1808–1886); Vincent A. Smith
(1848–1920); Heinrich Zimmer (1890–1943); and the extensive bibliography
of A. K. Coomaraswamy (1877–1947).
59. Several bilingual Sharada–Persian or Sharada–Arabic inscriptions have been
documented, though generally dating earlier: e.g., a foundation text from
the Tochi Valley (Khyber-Pakhtunkhwa Province, Pakistan), whose Kufic
Arabic portion provided the date of ah 243/847 ce, and the partially surviving
Sanskrit portion’s letters were “neither pure Nagari nor pure Sharada” (cf.
Dani et al., 1964: 128ff.); another bilingual Persian–Sharada inscription – the
Persian text in Kufic letters – was recovered near Zalamkot (Swat) and dated
to ah 401/1011 ce (see Rahman 1998).
60. Cf. Rehman and Hussain 2011: 69–70. I am also grateful to Dr. Jason Neelis
(pers. comm., April 2019) for sharing his insights on the Sharada script and its
regional histories. The relegation of this script to Kashmir may call for recon-
sideration: the inscription on this structure along with the recent rediscoveries
of Qarlugh coins with Sharada legends (second half of the thirteenth century),
minted at Ghazna and Kurraman, are collectively providing evidence of the
longer-lived and wider-spread of Sharada at least into the thirteenth century.
Cf. Bhandare 2020: 231–233.
61. See the discussion of this architectural configuration’s longevity in Ball 2020;
and Chapters 4 and 5.
62. With the systematic looting of historical sites, either for local building mate-
rials or for illicit trafficking of antiquities, structures change rapidly. Thus, as
Edwards conducted her documentation of the site during doctoral fieldwork
in the 1980s, she found more of the mihrab’s epigraphic program in situ than
I did more than ten years later in 1999.
63. Cf. supra in this chapter for a discussion of the textual references to possible
Ghaznavid architectural patronage in Lahore and Multan; and the documen-
tation of Ghaznavid traces in Swat.
64. Given the long presence of Muslim polities in the region – whether as elite
rulers or mercantile groups – there is little doubt that scholars of Islamic
theology and law were resident here and available for consultation. Cf. supra
in main text.
65. Cf. Chapter 4; Appendix, Afghanistan IV: 4, VI: 3.
66. See Chapter 5; Appendix, Afghanistan VIII: 4.
67. Cf. A. N. Khan 1988: 320–321; Ali 1993: 135; Flood 2001: 139–141.
68. See Appendix, Afghanistan III: 5, VIII: 4 and Pakistan III: 6, 20, 22.
69. For the riba†, see esp. Edwards 1990: 182–193, where connections are posited
between the choice of Qur’anic verses and the patron’s Karrami affinities;
also Edwards 1991: 92. Khan (1988: esp. 310–311) focused principally on the
foundation text, simply noting that the mihrab carried Qur’anic inscriptions
including the Throne verse (Q. II: 256), but this could not be verified by later
scholars (looting of fragments from the mihrab has been continuous). For
Ahmad Kabir, see Rehman and Hussain 2011: esp. 63–65.
70. E.g., Q. IX: 18 had previously appeared throughout the Isfahan oasis: around

the mihrab cupola of the Nayin jami‘ masjid (tenth century); on the Barsiyan
mihrab (ah 528/1134 ce); at the Muhammadiyya Mosque (ah 500/1106 ce);
and the jami‘s of Zavareh (ah 530/1136 ce) and Ardistan (ah 555/1160 ce).
Q. IX: 129 was also documented again at the Nayin jami‘. Cf. Dodd and
Khairallah 1981: II:43, 60; Blair 1992: 38, 177–178, 194.
71. Blair 1992.
72. Blair 1992: 206.
73. Cf. Husain 1936: 112, 113; Dodd and Khairallah 1981: II:113, 126, 149, 152; and
Cf. Rehman and Hussain 2011: 63.
74. A. N. Khan 1988: 311–316.
75. Edwards (2015: 190–200) considered the association of the saints with the
structures to be a later attempt at linking the site with better-known spiritual
figures (Shakar Ganj Shah is actually buried at Pakpattan), based on the first
appearance of the association being in J. W. Smyth’s Gazetteer of the Province
of Sind (1919). She referred to the structures primarily with their current
names, as also followed here.
76. See esp. the tomb of Yusuf Gardizi in Multan (Figure 6.15), which had been
established during the later eleventh century, but due to its maintenance as a
shrine and pilgrimage site over the subsequent centuries, the original, more
humble structure has developed into a tile-clad complex. Cf. Edwards 1990:
73–75; Rehman 1991: 43.
77. See esp. Patel 2004; Patel 2015.
78. Edwards 1990: 236ff.
79. See esp. Ali 1988: 51–56; Edwards 1990: 412ff.; Edwards 2006: 27–28.
80. Edwards (2015: 150) observed the presence of “glassy, colouristic effects” on
some Buddhist remains in the Indus valleys. But without excavated finds of
kilns in the vicinity of these pre-Islamic structures, it seems dubitable that the
decorations were of local production, possibly having been imported via the
numerous entrepôts of Sindh and Panjab.
81. See esp. Edwards 1990: 253.
82. Scerrato 1962.
83. Ali (1988: 44–45, pls. 31–34) assumed that some of the tile designs – particularly
those placed at doorframes – were undeciphered inscriptions, but his own
consultation of epigraphists seemed to disappoint this hope (cf. ibid., 96).
84. Azeem 1991: 90ff.
85. Deloche (trans. Walker) vol. I, 1993: 24–27. In contrast to the present-day
conditions – albeit among the Helmand Baluch, farther west in Afghan Sistan
– see Amiri’s (2020) description of “the dominance of the tribal system and
the lack of capital and trade routes.”
86. See esp. Pehrson 1966: 2; Spooner [1988] 2010: pt. 5.

Epilogue: Iran to India
T his volume has focused on an initially unknown and nomadic/tran-

shumant clan among the inhabitants of the region of Ghur, in what
is now central Afghanistan – their story beginning for our purposes in the
early twelfth century. During the course of this same century, roughly by
the 1150s ce, this group of Ghuri kinsmen emerged as a confederate empire
of multiple lineages. They adopted the trappings of Perso-Islamic king-
ship (Chapter 1) and called themselves Shansabanis. By the 1170s ce, the
three principal lineages of this confederacy were based at Jam-Firuzkuh,
Bamiyan, and the latest at the erstwhile Yamini capital of Ghazna. The
separate and at times contentious branches attained elite status and
regional dominance throughout what is modern Afghanistan, and tem-
porarily in eastern Iran and Turkmenistan. By the mid-1170s, the Ghazna
Shansabanis in particular seized upon that region’s geo-historical momen-
tum to expand into the Indus-dependent valleys, and ultimately into the
plains of the north Indian duab (Volume II).
This story of a meteoric rise from anonymity among central Afghanistan’s
nomad–urban continuum to the pinnacle of an empire encompassing a
variety of cultural landscapes was not as unlikely as it might first appear.
Indeed, by the twelfth century the Persianate world had experienced
the influx of several waves of “outsider” groups, particularly from the
eastern steppe lands of Turkic Eurasia. Chapter 1 traced how many of
these newcomers had been able not only to establish even more expansive
empires than the Shansabanis – the Saljuqs, for example – but also helped
transform the Persianate ecumene into one that was inclusive of lifeways
distinctly “other” to prior, conventional courtly models. The Persianate
world’s more aptly termed Perso-Islamic cultures of power mediated
access to kingship within this world by means of Islamic confessionalism,
patronage, and mobility. Rather than being exclusive and closed circuits
of prestige, then, Perso-Islamic kingly cultures were in fact attainable for
those groups and individuals with the perspicacity to decode them and the
means to implement their acquired knowledge.
Chapter 2 attempted to piece together specifically the Shansabanis’
beginnings among the nomads and other practitioners of transhumance in
central Afghanistan – an effort that remains incomplete given the general

Epilogue 339
paucity of textual and material traces of these populations through time.

The available data indicate nonetheless that their coalescence as an econom-
ically and then politically dominant regional presence was largely based on
increasing interactions with the mercantile communities throughout their
circuits of transhumance. Surely underpinning their ascendancy was the
generation of surplus resources, even while engaging in seasonal migra-
tions and pastoralism. During the eleventh and twelfth centuries – and
likely before – the ancestors of the Shansabanis had been exposed to
Islamic proselytism by variably heterodox groups such as the Karramiyya,
and quite likely Sufis and other spiritualists as well. But by the mid-twelfth
century, it was arguably the public acknowledgment and practice of an
overtly orthodox Islam that ultimately resulted in their emergence as a
regional and then transregional power, as well as the historical retrievabil-
ity of this process. The patronage of skilled labor to construct the recently
rediscovered Da Qadi minar encapsulated the Shansabanis’ twofold trans-
formation: regional politico-economic dominance and the concomitant
proclamation of an Islam-affiliated identity.
Subsequently, Chapter 3 traced the growing bifurcation in lifestyles
of the nominally senior or principal Shansabanis based at Firuzkuh, and
their agnates at Bamiyan, further highlighting the rapid – even alacritous –
adaptation of these two increasingly independent branches to their respec-
tive contexts. Despite Firuzkuh’s albeit tangential location within the
commercial networks radiating throughout Khurasan – networks that
also interwove Khurasan’s urban–intellectual centers – the Firuzkuh
Shansabanis appear to have remained tent-dwelling and continued their
seasonal migrations, essentially utilizing Firuzkuh as their sard-sir encamp-
ment or summer “capital.” The most recent archaeological work at the
site (2003–2005) revealed patronage of public structures, such as the more
famous minar and its adjacent mosque(s), rather than palaces. It was the
Shansabani branch at Bamiyan who exhibited palace-building and dwell-
ing, and perhaps increasing sedentism over time, as evidenced by archae-
ological surveys at Shahr-i Zuhak and Shahr-i Ghulghula in the Bamiyan
valley. This change of lifeways surely occurred in response to localized
requirements for the exercise and performance of kingship, entrenched in
the Bamiyan region’s cultural history.
As seen in Chapter 4, well before their declared espousal of the
Shafi‘iyya in ah 595/1199 ce, the Firuzkuh Shansabani elites had already
outgrown their exclusive loyalty to the Karramiyya: they extended their
contacts with the more orthodox – and politically and economically
more powerful – Khurasani madhhabs. The magnificent Shah-i Mashhad
madrasa (ah 560s/1165–1175 ce) demonstrated not only the Shansabanis’
emulation of their Ghaznavid predecessors in its epigraphic program,
the large and lavishly inscribed complex also hinted at their acuity in
diversifying patronage among the prominent legal–theological schools of

340 Iran to India
Khurasan. The Shansabanis had come a long way from their obscure and
humble beginnings among the nomads and transhumant groups of Ghur;
they were now imperial contenders, on the brink of establishing a trans-
regional empire.
Arguably, among the most pivotal moments in the Shansabanis’ impe-
rial arc was the conquest of Ghazna in ah 569/1173–1174 ce. But rather
than a centrally directed expansion of territorial presence, this conquest
brought about the establishment of a third Shansabani branch at the once
magnificent Yamini imperial capital. With the crowning of Shihab al-Din
as Sultan Mu‘izz al-Din of Ghazna, the Shansabani empire took on a
confederate yet centrifugal identity of hierarchically allied agnatic lineages,
which were always vying for greater independence. Indeed, the Ghazna
Shansabanis likely had the greatest momentum toward this independence,
given their access to new territorial conquests and economic resources to
the east in India. In any case, Chapter 5 explored the Shansabanis’ initial
architectural traces at Ghazna, which evinced a shift toward residence
in the previous rulers’ palaces: these were grandly renovated befitting
the new imperial occupants. Quite plausibly, the Ghazna Shansabanis’
incorporation of palace residency – alongside tent-dwelling – was both
an emulation of their defeated and displaced predecessors, and also a
response to politico-economic mores, this time pertaining in the region
of Zabulistan and incumbent on this new lineage if they were to exercise
effective imperial control. Overall, the mode of building empire that the
Ghazna Shansabanis evinced throughout Zabulistan and Zamindawar
could be termed re-inscription, wherein architectural complexes and their
surrounding areas were adopted under the aegis of a new rulership.
Following the eastward geo-political connections of Ghazna and
Zabulistan – forming a momentum impelling the Ghaznavids all the way
to the north Indian duab during the eleventh century – the Shansabanis
also looked toward the Indus and beyond for viable imperial expansion.
Chapter 6 examined the Shansabanis’ forays into the Indus-dependent
regions during the later 1170s and 1180s ce, which confronted them with
pre-existing imperial infrastructures established by the Ghaznavids in the
previous century. Re-inscription, their mode of imposing and exercising
imperial control amply evidenced in their initial conquests of the Ghaznavids’
core territories throughout Zabul–Zamindawar (Chapter 5), was applied on
exponentially larger scales: in the case of the Indus cities and their con-
tiguous cultural and economic landscapes, overlaying or re-inscribing the
Ghaznavids’ already existing imperial frameworks – for example, endowed
religious/“civic” institutions, coinage, circuits of mobility – with patronage
proclaiming Shansabani ascendancy was the most effective in capitalizing
on what was already there. Thus, the Ghazna Shansabanis were poised to
attempt an imperial formation not seen in the region for a thousand years:
the Ghaznavids could claim a largely peripatetic relationship with the north

Epilogue 341
Indian duab, but it had really been the Kushanas of the early centuries ce
who had last re-conjoined the Iranian and Indic cultural worlds for the
millennium to come. Despite a short imperial life of only about seven
decades, the Shansabanis – specifically the Ghazna lineage – can be credited
ultimately with leaving a substantial footprint in northern India. We are
prepared, then, for examination of the separate and yet intertwined histor-
ical processes unleashed by the Ghazna Shansabanis in northern India, and
their westward reverberations (Volume II).
Serving as a pivot between Volume I and Volume II is the question
aptly phrased by T. Barfield: “at what point should we stop treating [the
Shansabanis] as an example … of pastoral nomads and begin thinking of
them as elites with a nomadic pastoral heritage and history?”1 This query
is particularly germane, as the Shansabanis’ entire imperial activity took
place within a little more than half a century, and was thus within the
lifetimes of clansmen who, based on the localized circumstances of their
bases of power, effected transitions from more nomadic/transhumant to
more sedentist lifeways (cf. Chapters 3 and 4).
The fundamental question posed by Barfield may have multiple answers
as well as a singular, overarching one. The region-specific responses of the
three Shansabani lineages – shifting lifeways and economies as circum-
stances required – demonstrated the complex flexibility of the nomad–
urban continuum. It could also be said that non-elite empire-builders
became “elites with a nomadic pastoral heritage and history” when it
behooved them to think of themselves as such, desiring their court histori-
ans to project courtly images of them. And yet, the textual episteme of the
Persianate world afforded little conceptualization of non-courtly kingship,
historical reality being frequently sublimated via Persian historiographical
conventions (cf. Chapter 1). Ultimately, the ineluctable historical trace is
the Ghazna Shansabanis’ expansion into the Indus valleys and the north
Indian duab: for them, like their millennial predecessors the Kushanas (cf.
Introduction), nomadism and/or transhumance appeared to be inextricable
from their particular mode of empire-building, as it was their specifically
derived “cultural capital”2 (discussed in Chapters 1 and 5) that made such
a vast, transregional and transcultural imperialism possible. It could be the
case that the Shansabanis never fully transformed from nomadic or transhu-
mant pastoralists to “elites with a nomadic pastoral heritage and history.”
Notes
1. Barfield 2003.
2. Cf. Irons 2003.

Appendix: Shansabānī Religious and
Historical Inscriptions in Afghanistan
and Pakistan1
Introduction
This Appendix gathers the known Shansabānī-related epigraphy from

modern Afghanistan and Pakistan. Religious inscriptions refer to
Óadīth, Qur’anic verses, and doxologies; historical inscriptions contain
names, titles, dates, and other historical data. Overall, the inscriptions
gathered here date to the period of the Shansabānīs’ initial emergence
as politico-military elites in Afghanistan, specifically the three principal
agnatic lineages based at Jam-Fīrūzkūh, Bamiyan, and eventually Ghazna
spanning the third and fourth quarters of the twelfth century. The Ghazna
Shansabānīs’ rapid expansion into the lower Indus region – almost imme-
diately after the conquest of the erstwhile Yamīnī capital in ah 569/1173 ce
(see Chapter 5) – renders the inclusion of the more easterly epigraphic data
extremely informative. The published sources for the inscriptions are listed
in footnotes on each site, while their detailed discussion is to be found in
the relevant chapters of this volume.
Virtually since the beginning of European compilations of Islamic epig-
raphy for scholarly purposes – for example, the Répertoire Chronologique
d’Épigraphie Arabe (RCEA), initiated in 1931 at Cairo’s Institut Français
d’Archéologie Orientale – both religious and historical inscriptions have
been published together. This approach has implied parity in the practices of
documentation that are followed by the compilations, and has at least the-
oretically underscored the scholarly significance of both epigraphic genres.
However, this was in fact far from the case: as R. Hillenbrand2 has
pointed out specifically for the RCEA, historical inscriptions actually deter-
mined which religious inscriptions would be recorded, viz. only the reli-
gious texts preceding (very rarely following) historical ones were collected
(cf. also Chapter 4 notes). Without direct access to the sites of study,
investigators today would be largely working with only those inscriptions
that the editors of the RCEA and other compilations deemed worthy
of documentation. Fortunately, several scholarly works (books and arti-
cles) on epigraphy in various media – on architecture as well as portable
objects – have emerged, particularly in the latter half of the twentieth
century, to supersede such inherently biased methodologies.3

Appendix 343
The present Appendix aims to contribute to this ongoing effort, par-

ticularly for the twelfth century in Afghanistan and Pakistan: here are all
the known, deciphered, and discernible architectural inscriptions, con-
veying religious and/or historical content (e.g., Afghanistan IV: 2, 5–9).
Moreover, discussions of these findings in the relevant chapters of this
volume hopefully demonstrate the immense value of “purely” religious
inscriptions, even when they did not accompany historical epigraphic
content (cf. esp. Chapters 4 and 5).
It should be noted that, where inscriptions from periods other than
that of Shansabānī ascendancy are relevant, they are generally cited and
discussed in the notes (see, e.g., Pakistan III: 9). The one exception
to this practice is the inclusion of some Ghaznavid-period epigraphy
from Ghazna itself (cf. Afghanistan V and VI): the rationale guiding
this decision has been that Shansabānī architectural patronage, both at
this site and elsewhere in their territories, was undertaken in direct dia-
logue with the in situ monuments of their Ghaznavid predecessors. Since
Shansabānī-patronized monuments were to some degree responding to
those earlier architectural projects, the latter’s epigraphic programs are
provided here for reference.4
Finally, the inscriptions for each site and building have been arranged
according to the anticipated movement around and through the built
environment – essentially, a visitor’s experience of it – and the hypothet-
ical sequence of reading. Not only is it assumed that the Arabic–Persian
inscriptions would be read from right to left, they also follow consecutively
from exterior to interior spaces, and from the heights of elevations to their
bases (top down). In some examples, such as the small tombs in the vicin-
ity of Multan in the lower Indus region (Pakistan II and III), probable cir-
cumambulation would have directed a specific encounter of the epigraphy,
so that the order of inscriptions follows what was likely clockwise progress
around these structures.
Afghanistan
Balkh’s Western Vicinity, Juzjan Province
I. Tomb of Sālār Khalīl5

Exterior, main entrance
1. Surrounding entrance frame, Kufic, Arabic: Bismillah, identification

of structure as the tomb (‫ )مشهد‬of Sālār Khalīl.
2. Horizontal inset above entrance, foliated and knotted Kufic: shahāda.

344 Iran to India
Interior
3. Band around drum, bordered Kufic:

a. Bismillah; phrase from Qur’ān IV:170–172 (al-Nisa’); phrase
from Qur’ān XXVII: 19 (al-Naml); artisan Muªammad ibn
Aªmad ibn Maªmūd (6…‫)عمل محمد ابن احمد ابن محمود‬.
4. Continuous band following the trilobed arches, Kufic: ‫( الملك هلل‬peri-
odically ‫ الملك‬repeated).
5. Around the niche arches, floriated Kufic:
a. east niche arch: partial Qur’ān II:255 (al-Baqara);
b. north niche arch: Qur’ān II:163; Qur’ān X: 58 (Yūnus);
c. west niche arch: Qur’ān III:18–19 (Āl ‘Imrān).
Bust7
II. Arch8
1. Eastern face, Kufic:
a. right pier: Qur’ān II:127 (al-Baqara);
b. left pier incomplete historical inscription (Arabic):
9
...‫من اتمام هذه ]القبة؟[ في سنة‬
Chisht
III. Funerary Madrasa complex(?)10

Southwestern structure, exterior:
1. right of entrance, band framing entrance arch, naskh: indecipherable;

2. right of entrance, panel beneath arch, naskh: indecipherable;
3. left of entrance, panel beneath arch, naskh: Qur’ān XX:8 (˝aha).
Southwestern structure, interior:
4. Continuous horizontal band at springing of arches, naskh:

a. beginning in southeast corner, Arabic: bismillah; ren-
ovation (‫ )تجديد‬of the building dated to the reign
of Shams al-Dunya wa al-Dīn11; titles including
...‫الملك المعظم المؤيد المظفر‬
b. beginning in northeast corner, Arabic: royal titles, kunya
(‫)أبو الفتح‬, ism (‫)محمد‬, and na‚ab (‫;)ابن سام‬
c. beginning in northwest corner: Qur’ān LIX:23–24 (al-Óashr);
d. beginning in southwest corner: names of Allah.
5. Continuous band around arches, floriated Kufic:

Appendix 345
a. beginning in northeast corner: Qur’ān III:18–19 (Āl ‘Imrān);

b. beginning in northwest corner: Qur’ān II:255 (al-Baqara);
c. beginning in southwest corner: Qur’ān II:255–257 (al-Baqara);
d. beginning in southeast corner: Qur’ān II:257; Qur’ān CXII:1–4
(al-Ikhlā‚); Persian, ah 10 Jumādā I 562/4 March 1167 ce.12
Northeastern structure, exterior:
6. Band framing portal, bordered Kufic, Arabic–Persian:

a. right remnant: [‫السلطان المعظ]م‬
b. left remnant: ‫ بي تاريخ ربع االخر سنة‬...
c. springing of entrance arch:
i. north: ‫المعظم‬
ii. south: ‫السلطان‬
Northeastern structure, interior, west wall:
7. Large blind niche above right of miªrāb:

a. band delineating arch, floriated Kufic, Arabic: royal titles
including [‫]ا[العظم الملك ]؟[ رقاب األمم مو[لى ملوك العرب و العجم ؟‬
b. three (originally four?) medallions inside arch, naskh:
‫ علي‬،‫ عمر‬، ‫أبو بكر‬...
c. horizontal band below 7b, cursive: [‫و الحمد هلل]سبحان هللا‬
8. Large blind niche above left of miªrāb:
a. band delineating arch, floriated Kufic, Arabic: [‫ ]امير المؤمنين‬...
(termination of 7a);
b. four medallions inside arch, two legible, naskh:
...‫إسماعيل [؟] إبراهيم‬
c. horizontal band below 8b, termination of 7c, naskh:
‫و ال اله اال هللا و هللا ]ا[كبر‬
d. within medallion below 8c, floriated Kufic: ‫هللا‬
9. Smaller blind niches centered above miªrāb, floriated Kufic:
a. right, horizontal band below arch: ‫هللا اكبر‬
b. left, horizontal band below arch: ‫هللا اكبر‬
10. Small blind niche to left of 8:
a. upper band, Kufic: ‫هللا اكبر‬
b. inside panel, naskh: [‫عمل محمد ابن ابي بكر \ المعروف ]؟‬
Gharjistan
IV. Shah-i Mashhad madrasa remains13

South (principal) façade, central arch:

346 Iran to India
1. Knotted Kufic, Arabic: Bismillah; female patron; ah Ramadan 561 or

571/July 1166 or March 1176 ce.
South façade, central arch:
2. Lower edge of arch, nashkī, Arabic: reference to a male personage’s

grave:14 ‫نور هللا مرقده و يبصطه لحده‬
3. Inside central arch, thuluth: Takbīr of Tashrīq.15
South façade, right four arches:
4. Kufic, Qur’ān XLVIII:1–6 (al-Fatª);

5. Kufic, Arabic: Maªmūd, craftsman [?]: ...‫عمل محمود‬...; Aªmad ibn
Maªmūd, builder [?]‫تم هذه البنا العبد احمد ابن محمود‬...
North façade, extant west wall of entrance īwān:
6.Kufic, Bismillah;
7.Kufic, Qur’ān LIX:22–24 (al-Óashr);
8.Kufic: ‫ علي‬،‫ علي‬،‫( علي‬repetition);
9.Naskh, Óadīth enjoining Muslims “to be in the world like a
guest”;16
10. Naskh and thuluth, Arabic: ...‫رحل رسول هللا صلى هللا عليه و سلم‬
Ghazna17
V. Palace18
1. Top frieze running along approximately 148 marble dado slabs from
the Ghazna palace’s monumental north entrance; n.d., probably
eleventh–twelfth centuries (Ghaznavid); naskh, Arabic:
a. various benedictory texts in sequence.19
2. Three marble elements:20
a. upper half of small arch, naskh:
i. Qur’ān II: 256 (al-Baqara);
ii. laqab, kunya, and ism of Mas‘ūd III
(‫)السلطان العظم ابو سعد مسعود‬.
b. Transenna I, five bands around geometric grill, naskh:
i. ‫عمل محمد ابن حسين ابن مبارک‬
ii. …‫…و فرغ من اشدته‬
iii. ‫المبارک رمضان عظم هللا‬...‫في اول شهر‬
iv. ‫قدره سنه خمسين خمسمائه‬
v. ‫( خمس مائه‬ah 1 Rama∂ān 505/1–2 March 1112 ce].21
c. Transenna II, five bands around geometric grill, naskh:

Appendix 347
i. ‫ابن‬...‫عمل عثمان ابن‬

ii. …‫…و فرغ من اشدته‬
iii. ‫في اول شهر هللا‬
iv. ‫المبارک رمضان ]عظم[ هللا ]قدرته؟[ سنه‬
v. ‫[ خمس و خمس مايه‬ah 505/1112–1113 ce].22
3. Top frieze of marble dado slabs on walls of central court (partly in
situ), floriated Kufic, Persian: mathnavī (mutaqārib meter), approxi-
mately 85–95 distiches, progressing counterclockwise from northwest
through south and east sides; n.d., possibly early twelfth century
(r. Mas‘ūd III, 1099–1115 ce):23
a. west side:
i. …‫(…شاه محمود‬i.e., Maªmūd ibn Sabuktīgīn, r. ah 389–
421/999–1030 ce);24
ii. …‫( أبو سعيد‬i.e., Mas‘ūd I, r. ah 421–432/1030–41 ce).25
4. Top frieze of marble dado slabs on walls of central court (partly
in situ), floriated Kufic, Persian: qa‚īda [?] (mujtathth meter, rhyme
-an), progressing from northeast corner to northwest corner (possibly
extending further on both ends), likely two or more compositions.26
5. Baked brick and stucco fragments of monumental epigraphic bands,
recovered in west and northeast of courtyard (above marble dado
slabs V:1, 3):27
a. Kufic, Arabic: benedictory texts, e.g. ‫ الملک‬،‫( الملک هلل‬alone and/
or repeated);
b. Kufic: Bismillah; Qur’ān III:18–19 (Āl ‘Imrān) (stucco);28
c. three different fragments, naskh, Arabic: royal titles, including:
...‫( السلطان المعظم الم‬baked brick).29
VI. Minār of Mas‘ūd III30

1. Kufic: Bismillah; titles, laqab (‫ )السلطان العظم‬kunya (‫)ابو سعد‬, and ism
(‫ )مسعود‬of Sultan Mas‘ūd III.
2. Interlocking Kufic (in square panels): titles, kunya and ism of Mas‘ūd
III.
3. Naskh: Qur’ān XLVIII:1–29? (al-Fatª).
Ghūr
VII. Jam-Fīrūzkūh minār31

First (topmost) band:
1. Kufic: Shahāda.
Second band:
2. Kufic: Qur’ān LXI:13 and first three words of 14 (al-Íaf).

348 Iran to India
Third band:
3. Kufic, references to Ghīyāth al-Dīn:
...‫السلطان المعظم سلطان غياث الدنيا و الدين أبو الفتح محمد ابن سام‬
Fourth band:
4. Kufic, turquoise-blue tile letters: titles of Sul†ān Ghīyāth al-Dīn.
Bottom-most section, above base (tallest horizontal division with several

interspersed elements):
5. Undulating band forming circles and hexagons, Kufic: Qur’ān

XIX:1–98 (Maryām); on east face Qur’ān XIX:34 in stellate knot at
apex of large niche motif, probably serving as a miªrāb.32
6. Panel on north face:
a. craftsman’s signature, naskh (Arabic): ‫علي ابن ابراهيم النيشابوري‬
b. Kufic: ah 570/1174–1175 ce.
7. Octagonal socle, Kufic: titles of Sul†ān Ghīyāth al-Dīn (eroded).
Lashkari Bazar33
VIII. South or Great Palace (“Château du sud”)

Exterior of palace:
1. Great Mosque of Forecourt, around miªrāb:34

a. on flanking wall to north, Kufic: ‫( الملك هلل‬repeated; tentative);
b. on round pillars flanking north and south, Kufic, Arabic:
individual words, e.g., ‫ على‬,‫ امة‬,‫دولة‬, etc.;
c. among debris, originally on wall surfaces or pillars, Kufic: dox-
ologies, e.g., ‫ الملك هلل‬، ‫السلطان هلل العظمة هلل‬
2. South façade, Īvān X, Kufic, Arabic: probable renovation text dated
to ah 55x/1156–1165 ce.
Interior of palace:
3. Audience Hall I (northernmost īvān of palace, abutting and looking

out on Helmand):35
a. frieze around south entrance (partially in situ), Kufic:
i. Bismillah;
ii. Qur’ān XXVII:40–41 (al-Naml).
b. Brick and terracotta “ring-like” fragments of pillarets forming
part of overall decorative program, Kufic: ‫الملك هلل‬
4. Oratory F (southwest of Audience Hall I):
a. continuous frieze beginning on north wall and continuing

Appendix 349
on south wall, Kufic: Bismillah; Qur’ān XLVIII:1–6 or more

(al-Fa†h);
b. frieze framing miªrāb, Kufic: Bismillah; Qur’ān II:256
(al-Baqara);
c. upper part of miªrāb, Kufic: Qur’ān CXII:1–4 (al-Ikhlā‚)
repeated;
d. lower part of miªrāb, Kufic: doxologies, e.g.
‫ القدر هلل‬، ‫ السلطان هلل‬، ‫الملك هلل‬
Pakistan
Lower Indus Region (near Multan)
I. Ribā† of ‘Alī ibn Karmākh36

Miªrāb
1. First (outermost) vertical band, floriated Kufic: Qur’ān IX:18–19

(al-Tauba).
2. Second vertical band, floriated Kufic, founder/patron:
…‫…علي ابن کرماخ‬
3. Flanking vertical pilasters, floriated Kufic: Qur’ān LXI:13 (al-Íaf).
4. Hood, top three horizontal bands, Kufic: ‫( الملك هلل‬repeated).
5. Hood, bottom horizontal band, floriated Kufic: Qur’ān IX:129
(al-˝auba).
6. Squinches, Kufic: ‫( الملك‬right) ‫( هلل‬left).
7. Miªrāb back wall medallions: ‫ علي‬،‫ أبو بكر‬،‫ عمر‬، ‫ عثمان‬، ‫محمد‬
II. Tomb of “Sadan Shahid”37

Exterior, east façade:
1. Continuous vertical–horizontal–vertical band, from east–north to

east–south corner, naskh: Qur’ān XLVIII:1–3? (al-Fa†h).
2. Continuous vertical–horizontal–vertical band framing east–north
niche, Kufic: ‫( الملک هلل‬repeated).
3. Square projections above 2, naskh: ‫( هللا‬repeated).
4. East–north niche, gavāk‚a, naskh: ‫هللا‬
5. Continuous band framing entrance, naskh: ‫( هللا‬horizontally repeated,
alternating with floral insets).
6. Bases of pilasters flanking entrance, naskh: ‫هللا‬
7. Continuous vertical–horizontal–vertical band framing east-south

350 Iran to India
9. East–south niche, gavāk‚a, naskh: ‫هللا‬

10. Continuous horizontal band (above base moldings), naskh: ‫يا هللا‬
(repeated).
Exterior, south façade:
11. Continuous vertical–horizontal–vertical band framing southeast

niche: Kufic-style pseudo-epigraphy resembling ‫( الملک هلل‬repeated).
12. Ditto for southwest niche.
Exterior, north façade:
13. Continuous vertical–horizontal–vertical band framing northwest

15. Northwest niche, gavāk‚a, naskh: ‫هللا‬
16. Continuous band framing entrance: ‫( هللا‬horizontally repeated, alter-
nating with floral insets).
17. Continuous band framing northeast niche, Kufic: ‫( الملک هلل‬repeated).
19. Northeast niche, gavāk‚a, naskh: ‫هللا‬
20. Continuous horizontal band (above base moldings), naskh: ‫يا هللا‬
(repeated).
West façade:
21. Continuous vertical–horizontal–vertical band framing west–north

23. West–south niche, gavāk‚a, naskh: ‫هللا‬
24. Continuous band framing entrance, naskh: ‫( هللا‬horizontally repeated,
alternating with floral insets).
25. Bases of pilasters flanking entrance, naskh: ‫هللا‬
26. Continuous vertical–horizontal–vertical band framing west–south
28. West–north niche, gavāk‚a, naskh: ‫هللا‬
29. Continuous horizontal band (above base moldings), naskh:
‫( يا هللا‬repeated).
Interior, west upper wall:
30. Naskh, Arabic: beginning of two-line foundation inscription

(… ‫)امر بي بنا‬.

Appendix 351
III. Tomb of “Ahmad Kabir”38

East façade:
1. Socle medallions (two of four extant, right of stair [northeast]): swans.

2. East–north corner vertical band, thuluth: Qur’ān I:1 (al-Fātiªa).
3. Continuous band framing east–north niche, floriated Kufic: ‫الملک هلل‬
(repeated).
4. East–north niche panel, naskh: Qur’ān CVIII:1–3 (al-Kauthar).
5. Continuous band framing east–south niche, floriated Kufic: ‫الملک هلل‬
(repeated).
6. East–south niche panel, naskh: Qur’ān CXII:1–4 (al-Ikhlā‚).
7. East–south corner vertical band, thuluth: Qur’ān XLII:20 (al-Shū’rā).
South façade:
8. Socle medallions: swans (two).

9. Southeast niche panel: Śāradā (12 lines): … kalyā~a putrena …; Śaka
year 35; … Sitā …39
10. Southwest niche panel, “Western Kufic,”40 Arabic:
a. titles including ‫شهاب الدين…السلطان المعظم‬
b. ah 600/1203–1204 ce.
West façade:
11. Socle medallions (right to left): boar; horse; lion; elephant.

12. Vertical band, west–north corner: thuluth, Qur’ān I:1 (al-Fātiªa).
13. Parallel vertical band: thuluth (?), Qur’ān LV:2 (al-Raªmān).
14. West–south niche panel: naskh, Qur’ān CVIII:2–3? (al-Kauthar).
15. Vertical band, west–south corner: thuluth, Qur’ān LV:1 (al-Raªmān).
16. West–north niche panel: naskh, Qur’ān CVIII:1–2? (al-Kauthar).41
North façade:
17. Socle medallion (northeast extant): swan.

18. Northwest corner vertical band: thuluth, Qur’ān I:1 (al-Fātiªa).
19. Continuous band framing northwest niche: floriated Kufic, ‫الملك هلل‬
(repeated).
20. Northwest niche panel: naskh, Qur’ān CXII:1–2? (al-Ikhlā‚).
21. Continuous band framing northeast niche: floriated Kufic, ‫الملك هلل‬
(repeated).
22. Northeast niche panel: naskh, Qur’ān CXII:3–4? (al-Ikhlā‚).

352 Iran to India
Notes
1. Arranged alphabetically and chronologically.

2. Hillenbrand 2012: 19ff.
3. Among the most relevant works for Iran and Central Asia is of course
Blair’s (1992) comprehensive collection of inscriptions from these regions.
Additionally, R. Giunta’s masterly work on the funerary inscriptions found
at Ghazna (2003), along with the follow-up article of 2017, serve as important
sources for the inscriptions in Volume II (cf. also Afghanistan V–VI infra).
4. See esp. Chapters 4 and 5. A particularly noteworthy example of Ghaznavid–
Shansabānī imperial dialogue – or rather, the latter’s emulation of their
famous predecessors – comes from the Shansabānīs of Fīrūzkūh: their court
poet Fakhr al-Dīn Mubārakshāh al-MarvarrūÕī (d. ah 602/1206 ce) – to be
distinguished from the India-based poet known as Fakhr-i Mudābbir, a.k.a
Mubārakshāh (fl. mid-thirteenth century) – apparently composed one of only
two known pre-Mongol “dynastic mathnavīs,” quite possibly using as his
model the contents of the lengthy epigraphic program of the Ghazna palace’s
central court, which might have been the other surviving example of this
poetic genre (see Afghanistan V). Fakhr al-Dīn’s full composition is no longer
extant, known only through subsequent quotations from it, e.g., in Jūzjānī’s
˝abaqāt (vol. I, 1963: 318–319; trans. vol. I, 1873: 300–302). Cf. Khan 1977:
127; and esp. Allegranzi 2019a, vol. I: 143–144 and n. 82.
5. Also known as “Bābā Kha†īm ziyārat.” Cf. Melikian-Chirvani 1968; Sourdel-
Thomine 1971; Schneider 1984; and Introduction, above.
6. Sourdel-Thomine (1971: 315) characterized the Qur’anic references in this
inscription as “formules pieuses,” finding its remainder illegible. As later deci-
phered by Schneider (1984: 165), in fact the Qur’anic phrases appeared to
form part of a prayer for the compassion of Allah upon the named calligra-
pher/decorator or builder.
7. The present volume’s focus on the initial Shansabānī occupation of Bust–
Lashkari Bazar (see Chapter 5) has necessitated the separate treatment of
archaeological finds from sites: the well-known stelae recovered near Bust,
from the tomb locally attributed to a certain Shāhzāda Shaikh Óusain ibn
Shaikh Ibrāhīm, will be treated in Volume II. Nonetheless, see Sourdel-
Thomine 1956; Crane 1979; and Introduction, for a discussion of the Shāhzāda
Shaikh Óusain tomb.
8. See Sourdel-Thomine 1978: IB:65ff.; and especially Chapter 4.
9. Cf. Chapter 5 (including notes) for a discussion of the inscription’s variable
readings.
10. Cf. Blair 1985: 81–82; Kalus 2017: Fiche No. 37804; Ball 2019: No. 212. I reit-
erate my gratitude to Dr. Viola Allegranzi (2019b [forthcoming]) for sharing
her work in progress on Chisht’s epigraphy.
11. See Chapter 4 (esp. notes) for a discussion of the seemingly anachronistic
presence of this laqab: As of ah 558/1163 ce Bahā’ al-Dīn Sām’s elder son had
adopted Ghīyāth al-Dīn for his regnal title as sul†ān. See also Giunta and
Bresc 2004: 227.

Appendix 353
12. For a contextualization of the mixture of Arabic and Persian in the date, cf.
Giunta 2010: 177.
13. After Casimir and Glatzer 1971; Glatzer 1973; Blair 1985: 81; Najimi 2015. See
also Ball 2019: No. 1023.
14. Recorded only by Najimi (2015: 154); unfortunately, the author did not
include a photograph of this previously unread inscription.
15. Najimi (2015: 157, 167) proposed the identification of this takbīr; but see
Chapter 4.
16. Described as a specifically Shi‘ī Óadīth by Najimi (2015: 162–163), though
varying versions of it are to be found in Sunnī Óadīth collections as well – cf.
Chapter 4.
17. As was the case with Bust (cf. supra), this volume’s focus on the initial
expansion of the Shansabānīs beyond Ghūr has required that treatment of
the abundant funerary markers (principally tombstones) be undertaken in
Volume II (cf. also Giunta 2003a; Giunta 2003b; Giunta 2017).
18. Only forty-four slabs were excavated in situ, principally on the west but also
on the east and north dadoes of the courtyard; another 344 slabs were not
in situ. According to Bombaci’s (1966: 6) calculations based on the average
width, a total of about 510 slabs could have comprised the dado revetment of
the palace’s interior courtyard. The marble dado was placed below the baked
brick and stucco epigraphy above it (V: 5), probably datable to the Shansabānī
occupation of the palace. Cf. also Bombaci 1966: 33–36; Artusi 2009; Rugiadi
2009: 108ff.; Giunta 2010a: 164–166; Allegranzi 2019 (2 vols.); Allegranzi
2020a; Ball 2019: No. 358; Laviola 2020: 28–29.
19. These marble dado elements are to be distinguished from V:3–4, being asso-
ciated primarily with the northern entrance to the palace, or Hall XVII in the
excavated palace’s plan (see, e.g., Allegranzi 2019a, vol. I:71). Already in the
early seasons of excavations at the site, Scerrato (1959b, cited in Giunta 2010b)
had suggested they formed an independent epigraphic program, contained
within the north entrance alcove and separate from the larger compositions of
the central court. Cf. Rugiadi 2010: 2 (also cited in Allegranzi 2020a). Giunta
(2010b: 125) intriguingly observed that benedictory phrases – albeit with some
modifications – commonly occurred also on metalwork from eastern Iran
during the eleventh–thirteenth centuries. Cf. Laviola 2020: 456–457.
20. See esp. Bombaci 1966: 3, 19–20, figs. 131 and 133–136; Giunta and Bresc
2004: 171–172, 213–214; Giunta 2010a: 123–125; Kalus 2017: Fiche Nos. 36936,
36938, 43574, 43578, 43580, 43594, 43602, 43654. The rediscovery of these
fragments and their reconstitution as one element was paramount, as they
likely date the addition of the western oratory (XIII on the excavated palace’s
plan [Allegranzi 2019a, vol. I:71]) to the reign of Mas‘ūd III (r. ah 492–509/
1099–1115 ce).
21. The reading of Kalus (2017: Fiche No. 36936) is followed here; Giunta (2010b:
124) transliterated the date as corresponding to ‫خمس و خمس مايه‬.
22. This inscription does not figure among Kalus’s (2017; cf. note supra) re-
readings of the published texts, so that Giunta’s (2010b: 123) transliteration is
followed here.
23. Bombaci’s (1966: 13–15) initial epigraphic analysis of these extremely

354 Iran to India
important finds led him to observe that the meter(s) of the compositions on
the ex situ slabs was “not clear, but it seems not to contradict the mutaqārib or
the mujtathth of the text preserved in situ.” Allegranzi’s recent and extremely
meticulous study of the dado slabs provided a modified understanding of the
composition and contents of their inscriptions, as well as a period of compo-
sition probably post-mid-eleventh century, both of which are followed here.
Cf. also Allegranzi 2020a; and infra. For the slabs carrying the mutaqārib
meter, see Allegranzi 2019a, vol. I:70–84, 105–112, 120 and II:19–48, 64–76,
167–170, 173, 175–176, 183–192, 195–196, 204–207, 210–211, 213–216, 219.
Cited here (and in note infra) are all the dado fragments bearing a dis-
cernible meter, whether found in situ or recovered ex situ – including at
locales outside the Ghazna palace but in the city’s environs, some having
also been repurposed in other, usually funerary contexts (see, e.g., Laviola
2015).
24. Allegranzi’s (2019a, vol. I:119–120; II:29) analysis of several dado slabs in situ
along the central court’s west perimeter noted elegiac content, with reminis-
cences of the poet Farrūkhī’s (d. 1037–8?) compositions on the occasion of
Sul†ān Maªmūd’s death. It is noteworthy, perhaps, that the simple and direct
title of shāh is nowhere else associated with Sul†ān Maªmūd, absent among
other known numismatic and historical data; by contrast, it was associated
with the Shansabānī Sultan Mu‘īzz al-Dīn on an undated coin minted at
Ghazna (cf. Giunta and Bresc 2004: 166–168, 178, 206–208, 226).
25. Although very few certainties as to personages and dates emerge from the epig-
raphy of the Ghazna palace dadoes, the mention of Abū Sa‘īd (Sul†ān Mas‘ūd
I) and his posthumous title of Amīr-i Shahīd in the following distich is among
the few “incontestable” data, interpreted by Allegranzi (2019a, vol. I:120, 126
and II:36–37) as a definitive terminus post quem for the epigraphic project,
viz. 1041 ce or the death of Mas‘ūd I. See also Giunta and Bresc 2004: 168,
177, 208.
26. Cf. esp. Allegranzi 2019a, vol. I:114–115 and II:19–51, 61–63, 80–81, 83–84,
88–90, 92–96, 117, 123–124, 132–134, 159–160, 171–172, 193–195, 200–203,
217–218, 227–228, 233–235. The reader is referred to Allegranzi’s study for the
contents of the epigraphic fragments; again, only those lengthy enough to
convey meter are cited here (also noted supra).
27. A few of these fragments were photographed in situ during the Italian excava-
tions of the site in the 1950s–1960s. See Figures 5.6 and 5.7.
28. Not only were these Qur’anic verses extremely common in funerary inscrip-
tions), they are the only attested Qur’anic epigraphy in baked brick, which
has been dated to the Shansabānī occupation of the Ghazna palace in the
last quarter of the twelfth century. Cf. Giunta 2010a: 126; and Chapters 5
and 6.
29. While the laqab was primarily associated with Sul†ān Mu‘īzz al-Dīn after his
Ghazna victory (ah 568/1173–1174 ce) in numismatic, epigraphic, and textual
sources, it was also to be seen on several coins of Maªmūd ibn Ghīyāth al-Dīn
(r. ah 602–607/1206–1210 ce [Khwārazm-shāh vassal ah 604/1207–1208 ce]);
and on coins of three of the Shansabānī sul†āns of Bamiyan. Cf. Giunta and
Bresc 2004: 225.

Appendix 355
30. See Y. Godard 1936: I:367–369 and II:351; Pinder-Wilson 2001: 162–166;
Giunta and Bresc 2004: 177, 187, 193, 194, 198, 201.
31. After Sourdel-Thomine 2004; Giunta and Bresc 2004: 218, 225; Lintz 2013.
See also Ball 2019: No. 468.
32. See full discussion of the minār’s probable miªrāb and orientation (qibla) in
Chapter 4.
33. See esp. Sourdel-Thomine 1978: IB:11–15, 29–36, 42–50, 54–57 and planches;
also Allegranzi 2020a. The specific locations referred to within this site are
derived from Lashkari Bazar (1978) planches 2, 3, 4, 13, 23. Cf. also Ball 2019:
No. 685. The sequence of inscriptions listed here has taken into account
the probable dual principal axes of ingress into the palace: one from the
south (via the eponymous bazaar preceding the palace itself), which would
first bring the visitor to the forecourt’s mosque (VIII:1), and then onto the
main façade of the palace structure (VIII:2); and another, more ceremo-
nial entrance on the palace’s north perimeter from the Helmand’s banks
(VIII:3). The vastness of this site is well known, and to date its many struc-
tures (in varying states of preservation) have been only partially explored and
documented. The focus here is on decipherable epigraphic remains dated
either by content or style to Shansabānī presence; but several smaller architec-
tural ruins besides those deemed palaces were also attributed to this period,
based on anepigraphic motifs and their execution: e.g., Building Annex XX
(northeast of South Palace), and the so-called House of Racquets (Residence
XIII) about 1 km northeast of Bust. See Sourdel-Thomine, op. cit., 54–55,
60–61.
34. Although undated, Sourdel-Thomine (1978: IB:57) attributed the miªrāb’s
epigraphic program to a second phase of intervention here, namely,
Shansabānī renovations at the site overall (see also VIII:2–4), based on the
“refinement of certain forms of letters … that would be difficult to imagine
before the epigraphic development of the second half of the twelfth century
in Afghanistan” (my trans.).
35. Sourdel-Thomine (1978: IB:35–36) was in favor of a late Ghaznavid date for
parts of Audience Hall I’s epigraphic program, likening the execution to
works such as the minārs of Mas‘ūd III and Bahrām Shāh (cf. Afghanistan
VI). Even without Shansabānī renovations here, however, the ceremonial and
architectural predominance of this audience hall would have surely required
Mu‘īzz al-Dīn’s engagement with it after his victorious occupation of Lashkari
Bazar and its environs. See also Chapter 5.
36. Cf. Khan 1988: 307–377; Edwards 1990: 182–193; Edwards 1991: 92; Edwards
2015: 111–127, 201–210. The exterior of this structure is strikingly devoid of
ornament or epigraphy, at least partially due to renovations over the centu-
ries. Cf. Chapter 6.
37. See Khan 1988: 316–322; Ali 1993: 134–136; Flood 2001.
38. From Rehman and Hussain 2011.
39. Reproduced side-by-side with a line reading of the text in Rehman and
Hussain (2011: 69–70). Further examples of bilingual Śāradā–Persian or –
Arabic inscriptions have been published: cf. Dani et al., 1964; Rahman 1998;
and Chapter 6 notes.

356 Iran to India
40. This style of script was denoted by the authors (Rehman and Hussain 2011:
66, fig. 12), though without further explication.
41. It is possible that the epigraphic bands were rearranged in the course of the
intervening centuries: it is distinctly notable on this façade that the sequence
of verses from Qur’ān CVIII is reversed according to the process of circum-
ambulation. This reversal is to be contrasted to the north façade (Pakistan III:
20 and 22), on which the verses of Qur’ān CXII do appear according to the
progress of a circumambulating worshiper.

Works Cited and Bibliography
Historical Works
Anon. Tārīkh-i Sīstān, trans. M. Gold, Literary and Historical Texts from Iran, 2.
Roma: Istituto italiano per il Medio ed Estremo Oriente, 1976.
Bayhaqī, Abū al-FaÕl. The History of Beyhaqi (The History of Sultan Mas‘ud
of Ghazna, 1030–1041), trans. C. E. Bosworth, 3 vols. Boston, MA: Ilex
Foundation, 2011.
Gardīzī, ‘Abd al-Óayy. The Ornament of Histories: A History of the Eastern Islamic
Lands, AD 650–1041, ed. and trans. Clifford Edmund Bosworth, BIPS Persian
Studies Series. London: British Institute of Persian Studies and I. B. Tauris,
2011.
Jūzjānī, Abū ‘Amr Minhāj al-Dīn ‘Uthmān ibn Sirāj al-Dīn Muªammad.
A General History of the Muhammadan Dynasties of India, trans. Major H.
Raverty, Elibron Classics Replica, 2 vols. London: Gilbert & Rivington, 1881.
Jūzjānī, Abū ‘Amr Minhāj al-Dīn ‘Uthmān ibn Sirāj al-Dīn Muªammad.
˝abaqāt-i Nā‚iri, ed. Abdul Hai Habibi, 2nd edn. Kabul: Historical Society
of Afghanistan, 1963.
Samarqandī, NiÕāmī al-‘ArūÕī. Chahār Maqāla (The Four Discourses) of Nizami
al-Aruzi al-Samarqandi, trans. Edward G. Browne. 1978th edn. London: EJW
Gibb Memorial Trust, 1921.
Sanā’i, Óakīm Abū al-Majd Majdūd. The First Book of the Hadiqatu’l-Haqiqat,
or the Enclosed Garden of the Truth, trans. J. Stephenson. Calcutta: Baptist
Mission Press, 1910.
The Udana or the Solemn Utterances of the Buddha, trans. D. M. Strong. London:
Luzac, 1902, available at: https://archive.org/details/in.gov.ignca.9263.
Modern Persian Works
Abbārī, Asghar Farughī. Tārīkh-e Ghūrīyān (‫)تاريخ غوريان‬. Tehran: Sāzmān-e

Mu†āl‘a wa Tadwīn-e Kutub-e ‘Ulūm-e Insānī-ye Dāneshgāhā (SAMT), 2003.
Maªmūd, Maªmūd Shāh. Tārīkh-e impirā†ūrī-e Ghūrīyān (‫تاريخ امپراطوری‬
‫)غوريان‬. Kabul: Muʼassasa-ye Intishārāt-e Khāvar, 2009.
Pazkvāk, ‘A†īq Allah. Ghūrīyān (‫)غوريان‬. Kabul: Anjumān-i Tārīkh-i Afghānistān,
1968.
Roshanzamir, Mahdī. Tārīkh-e Sīyāsī wa NiÕāmī-yi Dūdmān-i Ghūrī
(‫)تاريخ سياسی و نظامی دودمان غوری‬. Tehran: Dāneshgāh-e Millī-e Irān, 1978.

358 Iran to India
Works on Afghanistan
Adamec, Ludwig W .(ed.). Historical and Political Gazetteer of Afghanistan, 6 vols.

Graz: Akademische Druck- u. Verlagsanstalt, 1972.
Adle, Chahryar. “Trois mosquées du début de l’ère islamique au Grand Khorassan:
Bastam, Noh-Gonbadan/Haji-Piyadeh de Balkh et Zuzan d’après des investi-
gations archéologiques,” in Greater Khorasan: History, Geography, Archaeology
and Material Culture (ed. Rocco Rante), 89–114. Berlin: De Gruyter,
2015.
Allegranzi, Viola. “Royal Architecture Portrayed in Bayhaqī’s TārīŸ-i Masʻūdī
and Archaeological Evidence from Ghazni (Afghanistan, 10th–12th c.).”
Annali dell’Istituto Orientale di Napoli 74 (2014): 95–120.
Allegranzi, Viola. “The Use of Persian in Monumental Epigraphy from Ghazni
(Eleventh–Twelfth Centuries).” Eurasian Studies 13 (2015): 23–41.
Allegranzi, Viola. Aux sources de la poésie ghaznavide. Les inscriptions persanes de
Ghazni (Afghanistan, XI–XII siècles), 2 vols. Paris: Presses Sorbonne Nouvelle,
2019a.
Allegranzi, Viola. “New Epigraphic Data from a Ghurid Monument in Chisht-i
Sharif: Expressing Power and Piety in 12th-century Afghanistan,” paper pre-
sented at Inscriptions from the Islamic World, Cairo, 6–8 September 2019b,
forthcoming.
Allen, Terry. “Notes on Bust.” Journal of the British Institute of Persian Studies,
Iran 26 (1988): 55–68.
Allen, Terry. “Notes on Bust (Continued).” Journal of the British Institute of
Persian Studies, Iran 28 (1990): 23–30.
Artusi, S. “Architectural Decoration from the Palace of Mas‘ud III in Ghazni:
The Italian Contribution (1957–2007),” in The IsIAO Italian Archaeological
Mission in Afghanistan 1957–2007: Fifty Years of Research in the Heart of Eurasia
(Proceedings of the Symposium Held in the Istituto Italiano per l’Africa e l’Oriente,
Rome, 8 January 2008), vol. 21: Conferenze (eds. Anna Filigenzi and Roberta
Giunta), 117–130. Rome: Istituto Italiano per l’Africa e l’Oriente, 2009.
Azad, Arezou. Sacred Landscape in Medieval Afghanistan: Revisiting the Fa∂ā’il-i
Balkh. Oxford: Oxford University Press, 2013.
Baker, P. H. B. and F. Raymond Allchin. Shahr-i Zohak and the History of the
Bamiyan Valley, Afghanistan, Ancient India and Iran Trust Series, No. 1.
Oxford: Tempvs Reparatvm, 1991.
Ball, Warwick. “The Seven Qandahars: The Name Q.ND.HAR. in the Islamic
Sources.” South Asian Studies 4 (1988): 115–142.
Ball, Warwick. “Chapter 12: General Conclusions: Historical Overview of the
Kandahar Sequence,” in Excavations at Kandahar 1974 and 1975 (eds. Anthony
McNicoll and Warwick Ball), 391–402. Oxford: Tempus Reparatum, 1996.
Ball, Warwick. “The Towers of Ghur: A Ghurid ‘Maginot Line’?” in Cairo to
Kabul (eds. Warwick Ball and Leonard Harrow), 21–45. London: Melisende,
2002.
Ball, Warwick. The Monuments of Afghanistan: History, Archaeology and
Architecture. London: I. B. Tauris, 2008.
Ball, Warwick. “Afghanistan, Art and Architecture.” In Encyclopedia of Islam,

3rd edn (eds. Gudrun Krämer, Denis Matringe, John Nawas, and Everett
Rowson). Leiden: Brill Online, 2013.
Ball, Warwick. Archaeological Gazetteer of Afghanistan, revised edn. Oxford:
Oxford University Press, 2019.
Ball, Warwick. “Buddhist Elements in the Architecture of Afghanistan, 1000–
1250,” in The Architecture of the Greater Iranian World, 1000–1250 (ed. Robert
Hillenbrand). Edinburgh: Edinburgh University Press, 2020.
Ball, Warwick, Olivier Bordeaux, David W. MacDowall, Nicholas Sims-Williams,
and Maurizio Taddei. “Chapter 6. From the Kushans to the Shahis,” in The
Archaeology of Afghanistan from Earliest Times to the Timurid Period (eds. F.
R. Allchin, Norman Hammond, and Warwick Ball), 344–459. Edinburgh:
Edinburgh University Press, 2019.
Ball, Warwick and Yolanda Crowe. “General Conclusions: Historical Overview
of the Kandahar Sequence,” in Excavations at Kandahar 1974 and 1975 (eds.
Anthony McNicoll and Warwick Ball), 391–402. Oxford: Tempus Reparatum,
1996.
Ball, Warwick and Klaus Fischer. “Chapter 7. From the Rise of Islam to the
Mongol Invasion,” in The Archaeology of Afghanistan from Earliest Times to the
Timurid Period (eds. F. R. Allchin, Norman Hammond, and Warwick Ball),
460–545. Edinburgh: Edinburgh University Press, 2019.
Ball, Warwick, Simon Glenn, Bertille Lyonnet, David W. MacDowall, and
Maurizio Taddei. “Chapter 5. The Iron Age, Achaemenids, and Hellenistic
Periods,” in The Archaeology of Afghanistan from Earliest Times to the Timurid
Period (eds. F. R. Allchin, Norman Hammond, and Warwick Ball), 260–343.
Edinburgh: Edinburgh University Press, 2019.
Bernard, Paul and Frantz Grenet. “Découverte d’une statue du dieu solaire Surya
dans la région de Caboul.” Studia Iranica 10(1) (1981): 127–146.
Bernardini, Michele. “Les Qaraunas à Ghazni, XIIIe–XIVe siècles,” in Texts and
Contexts: Ongoing Researches on the Eastern Iranian World (Ninth–Fifteenth C.)
(eds. Viola Allegranzi and Valentina Laviola), 249–268. Rome: Istituto per
l’Oriente C. A. Nallino, 2020.
Bivar, A. D. H. “Seljuqid Ziayarats of Sar-i Pul (Afghanistan).” Bulletin of the
School of Oriental and African Studies 29(1) (1966): 57–63.
Bivar, A. D. H. “Review of Lashkari Bazar: une résidence royale ghaznévide et
ghoride.” Bulletin of the School of Oriental and African Studies 43(2) (1980):
385–386.
Bombaci, Alessio. “Ghazni.” East and West 8(3) (1957): 247–260.
Bombaci, Alessio. The Kūfic Inscription in Persian Verses in the Court of the Royal
Palace of Masʻūd III at Ghazni, Reports and Memoirs, 5. Rome: IsMEO, 1966.
Bosworth, C. E. “Ghaznevid Military Organisation.” Der Islam 36(1/2) ([1960]
1977): 37–77.
Bosworth, C. E. “The Early Islamic History of Ghur.” Central Asiatic Journal 6(2)
(1961): 116–133.
Bosworth, C. E. The Ghaznavids, Their Empire in Afghanistan and Eastern Iran
(994–1040). Beirut: Librarie du Liban, [1963] 2015.
Bosworth, C. E. “Notes on the Pre-Ghaznavid History of Eastern Afghanistan.”
The Islamic Quarterly: A Review of Islamic Culture 9 (1965): 12–24.

360 Iran to India
Bosworth, C. E. “Mahmud of Ghazna in Contemporary Eyes and in Later Persian

Literature.” Iran IV (1966): 85–92.
Bosworth, C. E. Sistan under the Arabs, from the Islamic Conquest to the Rise of the
Saffarids (30–250/651–864). Rome: Istituto Italiano per il Medio ed Estremo
Oriente, 1968a.
Bosworth, C. E. “The Coming of Islam to Afghanistan,” in Islam in Asia (ed.
Yohanan Friedmann), 1:1–22. Boulder, CO: Westview Press, 1984.
Bosworth, C. E. “Balkh ii. History from the Arab Conquest to the Mongols,” in
Encyclopaedia Iranica, online edition, 1988.
Bosworth, C. E. The Later Ghaznavids: Splendour and Decay, 1st Indian edn. New
Delhi: Munshiram Manoharlal Publishers, [1977] 1992.
Bosworth, C. E. The History of the Saffarids of Sistan and the Maliks of Nimruz.
Costa Mesa, CA: Mazda, 1994.
Bosworth, C. E. “Firuzkuh,” in Encyclopaedia Iranica, December 15, 1999.
Bosworth, C. E. “The Appearance and Establishment of Islam in Afghanistan,”
in Islamisation de l’Asia Centrale (ed. Étienne de la Vassière), 97–114. Paris:
Association pour l’avancement des études iraniennes, 2008.
Bosworth, C. E. “Additions to the New Islamic Dynasties,” in Living Islamic
History: Studies in Honour of Carole Hillenbrand (ed. Y. Suleiman), 14–31.
Bosworth, C. E. “Fakhr-i Mudabbir,” in Encyclopaedia of Islam, 2nd edn (eds. Th.
Bianquis, C. E. Bosworth, E. van Donzel, W. P. Heinrichs, and P. Bearman).
Leiden: Brill online, 2012a.
Bosworth, C. E. “Karramiyya,” in Encyclopedia of Islam, 2nd edn (eds. P. Bearman,
Th. Bianquis, C. E. Bosworth, E. van Donzel, and W. P. Heinrichs). Leiden:
Brill online, 2012b.
Bosworth, C. E. “Ghurids,” in Encyclopaedia Iranica, online edition, 2012c.
Bosworth, C. E. “Ghurids,” in Encyclopaedia of Islam, 2nd edn (eds. P. Bearman,
Th. Bianquis, C. E. Bosworth, E. van Donzel, and W. P. Heinrichs). Leiden :
Brill online, 2014.
Casimir, Michael and Bernt Glatzer. “Shah-i Mashhad, a Recently Discovered
Madrasah of the Ghurid Period in Gargistan.” East and West 21(1/2) (1971):
53–68.
Crane, Howard. “Helmand–Sistan Project: An Anonymous Tomb at Bust.” East
and West 29 (1979): 241–246.
Cribb, Joe. “Chapter Four: The Monetary History of Begram,” in Charles
Masson: Collections from Begram and Kabul Bazaar, Afghanistan, 1833–
1838 (ed. Elizabeth Errington), 79–206. London: The British Museum,
2020.
Crowe, Yolanda. “Chapter 10: Glazed Ceramics,” in Excavations at Kandahar
1974 and 1975 (eds. Anthony McNicoll and Warwick Ball), 313–364. Oxford:
Tempus Reparatum, 1996.
Dagens, Bruno, Marc Le Berre, and Daniel Schlumberger. Monuments préislam-
iques d’Afghanistan, Mémoires de la Délégation Archéologique Française en
Afghanistan, vol. 19. Paris: Librairie G. Klincksieck, 1964.
de la Vassière, Étienne. “Inherited Landscapes in Muslim Bactra.” Eurasian
Studies 16 (2018): 124–141.

de Planhol, X. “Bamian i. The Bamian Basin,” in Encyclopaedia Iranica, online

edition, 1988.
de Planhol, X. “Gazni i. Historical Geography,” in Encyclopaedia Iranica, online
edition, [2000] 2012.
Dupree, Louis. Afghanistan, 8th edn. Karachi: Oxford University Press, [1997 3rd
edn.] 2012.
Dupree, Nancy Hatch. An Historical Guide to Afghanistan, 2nd edn. Kabul:
Afghan Tourist Organization, 1977.
Errington, Elizabeth. Charles Masson and the Buddhist Sites of Afghanistan:
Explorations, Excavations, Collections 1832–1835, Research Publication, No. 215.
London: The British Museum Press, 2017.
Fischel, Walter J. “The Rediscovery of the Medieval Jewish Community at
Fīrūzkūh in Central Afghanistān.” Journal of the American Oriental Society
85(2) (1965): 148–153.
Fischer, Klaus. “Archaeological Remains of Jainism in West Pakistan and
Afghanistan.” Voice of Ahimsa 121 (1962).
Fischer, Klaus. “Preliminary Remarks on Archaeological Survey in Afghanistan.”
Zentralasiatische Studien 3 (1969): 327–408.
Fischer, Klaus. “Projects of Archaeological Maps from Afghan-Seistan between
3120ʹ to 3050ʹN and 6200ʹ to 6210ʹE (with Three Maps in Folder).”
Zentralasiatische Studien 4 (1970): 483–534.
Fischer, Klaus. “Historical, Geographical and Philological Studies on Seistan by
Bosworth, Daffinà and Gnoli in the Light of Recent Archaeological Field
Surveys.” East and West 21 (1971): 45–51.
Fischer, Klaus. “Chapter 6: From the Rise of Islam to the Mongol Invasion,” in
The Archaeology of Afghanistan from Earliest Times to the Timurid Period (eds.
F. R. Allchin and Norman Hammond), 301–355. New York: Academic Press,
1978.
Fischer, Klaus, Dietrich Morgenstern, and Volker Thewalt. Geländebegehungen in
Sistan 1955–1973 Und Die Aufnahme von Dewal-i Khodaydad 1970, vol. 2: Plates,
Maps, Plans, 2 vols. Bonn: Rudolf Habelt Verlag, 1974.
Flood, Finbarr B. “Between Cult and Culture: Bamiyan, Islamic Iconoclasm, and
the Museum.” The Art Bulletin 84(4) (2002): 641–659.
Flood, Finbarr B. “Ghurid Monuments and Muslim Identities: Epigraphy and
Exegesis in Twelfth-Century Afghanistan.” Indian Economic and Social History
Review 42(3) (2005a): 263–294.
Flood, Finbarr B. “Review of Le minaret ghouride de Jam: un chef d’oeuvre du XIIe
siècle by J. Sourdel-Thomine.” Art Bulletin 87(3) (2005b): 536–543.
Flood, Finbarr B. “Islamic Identities and Islamic Art: Inscribing the Qur’an
in Twelfth-Century Afghanistan.” Studies in the History of Art 74 (2009b):
91–117.
Flury, S. “Le décor épigraphique des monuments de Ghazna.” Syria (Revue d’Art
Oriental et d’Archéologie) 6 (1925): 61–90.
Franke, Ute. “Ancient Herat Revisited. New Data from Recent Archaeological
Fieldwork,” in Greater Khorasan: History, Geography, Archaeology and Material
Culture (ed. Rocco Rante), 63–88, Studies in the History and Culture of the
Middle East. Berlin: De Gruyter, 2015.

362 Iran to India
Franke, Ute. “Glazed Earthenware from the 10th to the 13th Century,” in Herat
through Time, vol. 3: Ancient Herat (eds. Ute Franke and Martina Müller-
Wiener), 138–183. Berlin: Staatliche Museen zu Berlin, 2016a.
Franke, Ute. “Herat from the 10th to the 14th Century,” in Herat through Time,
vol. 3: Ancient Herat (eds. Ute Franke and Martina Müller-Wiener), 75–86.
Berlin: Staatliche Museen zu Berlin, 2016b.
Franke, Ute. “Monochrome Fritware from the 12th and Early 13th Century,” in
Herat through Time, vol. 3: Ancient Herat, 359–371. Berlin: Staatliche Museen
zu Berlin, 2016c.
Franke, Ute. “Monochrome Glazed Earthenware,” in Herat through Time, vol. 3:
Ancient Herat (eds. Ute Franke and Martina Müller-Wiener), 319–357. Berlin:
Staatliche Museen zu Berlin, 2016d.
Franke, Ute. “Unglazed Pottery from the 10th to the 13th Century: Magic Motifs,”
in Herat through Time, vol. 3: Ancient Herat, 231–317. Berlin: Staatliche Museen
zu Berlin, 2016d.
Franke, Ute. “Archaeological Research in Qal’a-e Ekhtyaruddin: Excavations in
the Upper Citadel – Trenches 1a and 1b,” in Excavations and Explorations in
Herat City, vol. 2: Ancient Herat (eds. Ute Franke and Thomas Urban), 93–319.
Berlin: Staatliche Museen zu Berlin, 2017.
Franke, Ute. “Résumé: New Perspectives on Ancient Herat,” in Excavations and
Explorations in Herat City, vol. 2: Ancient Herat (eds. Ute Franke and Thomas
Urban), 743–751. Berlin: Staatliche Museen zu Berlin, 2017.
Franke, Ute, Benjamin Mutin, and Cécile Buquet-Marcon. “Excavations in
Kuhandaz,” in Excavations and Explorations in Herat City, vol. 2: Ancient Herat
(eds. Ute Franke and Thomas Urban), 689–731. Berlin: Staatliche Museen zu
Berlin, 2017.
Franke, Ute and Thomas Urban. “Areia Antica – Ancient Herat: Summary of the
Work Carried out by the DAI-Mission in Collaboration with the Institute of
Archaeology, Ministry of Information and Culture, Kabul, August–September
2006.” Berlin: German Foreign Office; German Archaeological Institute
(DAI), 2006.
Franklin, Kathryn and Emily Boak. “The Road From Above: Remotely
Sensed Discovery of Early Modern Travel Infrastructure in Afghanistan.”
Archaeological Research in Asia 19 (2019): 40–54.
Frye, Richard N. and Clifford Edmund Bosworth. “Herat,” in Historic Cities of
the Islamic World (ed. C. E. Bosworth), 153–155. Leiden: Brill, 2007.
Gardin, J-C. “Poteries de Bamiyan.” Ars Orientalis 2 (1957): 227–245.
Gardin, J-C. Lashkari Bazar: une résidence royale ghaznévide et ghoride, vol. 2: Les
trouvailles; céramiques et monnais de Lashkari Bazar et de Bust, 2 vols. Paris:
Klincksieck, 1963.
Gascoigne, Alison. “Pottery from Jam: A Medieval Ceramic Corpus from
Afghanistan.” Iran 48 (2010): 107–151.
Ghafur, Muhammad Abdul. “The Gorids: History, Culture and Administration,”
University of Hamburg, 1960.
Giunta, Roberta. “Some Brief Remarks on a Funerary Stele Located in the Gazni
Area (Afghanistan).” East and West 51(1/2) (2001): 159–165.
Giunta, Roberta. Les inscriptions funeraires de ˛aznī: (IVe–IXe/Xe–XVe Siècles),

Series Maior, 8. Napoli: Università degli studi Napoli “L’orientale”,

Dipartimento di studi asiatici, 2003a.
Giunta, Roberta. “Un texte de construction d’époque guride à Gazni,” in Studi in
Onore de Umberto Scerrato per Il Suo Settantacinquesimo Compleanno, 439–455.
Rome, 2003b.
Giunta, Roberta. “Islamic Ghazni An IsIAO Archaeological Project in Afghanistan:
A Preliminary Report (July 2004–June 2005).” East and West 55(1/4) (2005):
473–484.
Giunta, Roberta. “A Selection of Islamic Coins from the Excavations of Udegram,
Swat.” East and West 56(1/3) (2006): 237–262.
Giunta, Roberta. “Les inscriptions persanes dans l’épigraphie monumentale de
la ville de Ghazni (Afghanistan) aux 6e–7e/12e–13e siècle,” in Iranian Identity
in the Course of History. Proceedings of the Conference held in Rome, 21–24
September 2005 (ed. Carlo G. Cereti), 163–180, Serie Orientale Roma CV.
Rome: Istituto Italiano per l’Africa e l’Oriente, 2010a.
Giunta, Roberta. “New Epigraphic Data from the Excavations of the Ghaznavid
Palace of Mas‘ud III at Ghazni (Afghanistan),” in South Asian Archaeology
2007: Proceedings of the 19th Meeting of the European Association of /South Asian
Archaeology, Ravenna, Italy, July 2007 (eds. Pierfrancesco Callieri and Luca
Colliva), BAR International Series 2133, vol. 2, 122–131. Oxford: Archaeopress,
2010b.
Giunta, Roberta. “Gazni ii. Monuments and Inscriptions,” in Encyclopaedia
Iranica, online edition, [2000] 2012.
Giunta, Roberta. “Tombeaux et inscriptions funeraires de Ghazni (Afghanistan).
Quelques documents inédits du XIe–XIIIe siècle.” Vicino Oriente XXI (2017):
127–145.
Giunta, Roberta. “Les études sur la documentation archéologique et épigraphique
de Ghazni: Résultats et nouvelles pistes de recherche,” in Texts and Contexts:
Ongoing Researches on the Eastern Iranian World (Ninth–Fifteenth C.) (eds.
Viola Allegranzi and Valentina Laviola), 157–186. Rome: Istituto per l’Oriente
C.A. Nallino, 2020.
Giunta, Roberta and Cécile Bresc. “Listes de la titulature des ghaznavides et des
ghurides à travers les documents numismatiques et épigraphiques.” Eurasian
Studies III(1) (2004): 161–243.
Glatzer, Bernt. “The Madrasah of Shah-i Mashhad in Badgis.” Afghanistan 25(4)
(1973): 46–68.
Glatzer, Bernt. “Das Mausoleum und die Moschee des Ghoriden Ghiyat-ud-Din
in Herat.” Afghanistan Journal 7(1) (1980): 6–22.
Grenet, Frantz. “The Nomadic Element in the Kushan Empire (1st–3rd Century
ad).” Journal of Central Eurasian Studies 3 (2012): 1–22.
Habibi, A. H. “The City of Ferozkoh: Where Was It?” Afghanistan 33(1) (1980):
34–44.
Hackin, Joseph and Johann Carl. Nouvelles Rechèrches Archéologiques à Bamiyan,
vol. III : Mémoires de la délégation archéologique française en Afghanistan. Paris:
Les Éditions G. Van Oest, 1933.
Haim, Ofir. “An Early Judeo-Persian Letter Sent from Ghazna to Bāmiyān (Ms.
Heb. 4°8333.29).” Bulletin of the Asia Institute 26 (2012): 103–120.

364 Iran to India
Haim, Ofir. “What Is the ‘Afghan Genizah’? A Short Guide to the Collection of
the Afghan Manuscripts in the National Library of Israel, with the Edition of
Two Documents.” Afghanistan 2(1) (2019): 70–90.
Hammer, Emily, Rebecca Seifried, Kathryn Franklin, and Anthony Lauricella.
“Remote Assessments of the Archaeological Heritage Situationin Afghanistan.”
Journal of Cultural Heritage, 2017, 20 pp.
Hansen, Erik, Abdul Wasay Najimi, and Claus Christensen. The Ghurid Portal
of the Friday Mosque of Herat, Afghanistan. Aarhus: Aarhus University Press,
2015.
Helms, S. W. “Excavations at ‘The City and the Famous Fortress of Kandahar, the
Foremost Place in All of Asia.’” Afghan Studies 3/4 (1982): 1–24.
Helms, S. W. “Kandahar of the Arab Conquest.” World Archaeology 14(3) (1983):
343–353.
Herberg, Werner. “Das Land Ghor in Afghanistan: auf der Suche nach einem
verschollen Imperium.” Die Waage 17(5) (1978): 216–220.
Herberg, Werner. “Die Wehrbauten von Ghor (Afghanistan): Zusammenfassende
Dokumentation Der Bestandsaufnahmen von 1975, 1977 Und 1978.” Die Welt
Des Islams N.S. 22(1/4) (1982): 67–84.
Herberg, Werner and D. Davary. “Topographische Feldarbeiten in Ghor: Bericht
über Forschungen zum Probleme Jam-Ferozkoh.” Afghanistan Journal 3(2)
(1976): 57–69.
Hillenbrand, Robert. “The Architecture of the Ghaznavids and Ghurids,” in The
Sultan’s Turret: Studies in Persian and Turkish Culture. Studies in Honor of
Clifford Edmund Bosworth (ed. Carole Hillenbrand), vol. 2, 124–206. Leiden:
Brill, 2000.
Hillenbrand, Robert. “The Ghurid Tomb at Herat,” in Cairo to Kabul (eds.
Warwick Ball and Leonard Harrow), 123–143. London: Melisende, 2002.
Hunter, E. C. D. “Hebrew-Script Tombstones from Jam, Afghanistan.” Journal
of Jewish Studies 61(1) (2010): 72–87.
Inaba, Minoru. “Kandahar iii. Early Islamic Period,” in Encyclopaedia Iranica,
online edition, 2010.
Inaba, Minoru. “Chapter Two: Sedentary Rulers on the Move: The Travels of the
Early Ghanzavid Sultans,” in Turko-Mongol Rulers, Cities and City Life (ed.
David Durand-Guédy), 75–98. Leiden: Brill, 2013.
Inaba, Minoru. “The Narratives on the Bāmiyān Buddhist Remains in the Islamic
Period,” in Encountering Buddhism and Islam in Premodern Central and South
Asia (eds. B. Auer and I. Strauch), 75–96 (draft). Berlin: De Gruyter, 2019.
Jackson, Peter. “The Fall of the Ghurid Dynasty,” in The Sultan’s Turret: Studies
in Persian and Turkish Culture. Studies in Honor of Clifford Edmund Bosworth
(ed. Carole Hillenbrand), 2:208–235. Leiden: Brill, 2000.
Johnstone, T. M. “Ghaznawids.” in Encyclopaedia of Islam, vol. II. Leiden: Brill,
1965.
Khan, M. S. “The Life and Works of Fakhr-i-Mudabbir.” Islamic Culture LI
(1977): 127–140.
Khazeni, Arash. “Fortressed: The Face of the Durrani Kingdom of Afghanistan,
circa 1817,” paper presented at Twelfth Biennial Iranian Studies Conference,
University of California, Irvine, August 2018.

Klimburg-Salter, Deborah. The Kingdom of Bamiyan: Buddhist Art and Culture

of the Hindu Kush, Series Maior, 5. Naples: Istituto universitario orientale,
Dipartimento di studi asiatici, 1989.
Klimburg-Salter, Deborah. “Buddhist Painting in the Hindu Kush, ca. VIIth to
Xth Centuries: Reflections on the Co-Existence of Pre-Islamic and Islamic
Artistic Cultures during the Early Centuries of the Islamic Era,” in Islamisation
de l’Asia Centrale (ed. Étienne de la Vassière), 131–159. Paris: Association pour
l’avancement des études iraniennes, 2008.
Kohzad, A. A. (Ahmed Ali). “Along the Koh-i Baba and Hari-Rud.” Afghanistan
6(1) (1951a): 1–16.
Kohzad, A. A. “Along the Koh-i Baba and Hari-Rud. Part II.” Afghanistan 6(2)
(1951b): 1–21.
Kohzad, A. A. “Along the Koh-i Baba and Hari-Rud. Part III.” Afghanistan 7(1)
(1952): 50–55.
Kohzad, A. A. “Along the Koh-i Baba and Hari-Rud. Part IV.” Afghanistan 8(4)
(1953): 54–65.
Kohzad, A. A. “Along the Koh-i Baba and Hari Rud. Part V.” Afghanistan 9(1)
(1954a): 20–43.
Kohzad, A. A. “Along the Koh-i Baba and Hari-Rud. Part VI.” Afghanistan 9(2)
(1954b): 1–21.
Kohzad, A.A. “Afghanistan: Geographical and Historical Sketches of Some
Localities.” East and West 7(2) (1956): 128–137.
Kohzad, A. A. “Firoz Koh.” Afghanistan 12(4) (1957): 31–34.
Kohzad, M. N. (Mohammed Nabi). “Tourist Sites of Afghanistan.” Afghanistan
14(3) (1959): 1–7.
Kristy, Gwendolyn. “The Impact of Urban Sprawl on Cultural Heritage in Herat,
Afghanistan: A GIS Analysis.” Digital Applications in Archaeology and Cultural
Heritage 10 (2018): 8 pp.
Kufi, Ali. The Chachnamah, An Ancient History of Sind, trans. Mirza Kalichbeg
Fredunbeg. Delhi: Idarah-i Adabiyat-i Delli, 1900.
Laviola, Valentina. “Ziyarat of Ghazni: Three Case Studies about Marble
Re-Employments.” Eurasian Studies 13 (2015): 42–53.
Laviola, Valentina. “Starting the Analysis of Stucco Finds from Ghazni Royal
Palace (11th–12th c.),” paper presented at Textes et Contextes: Rechèrches en
cours sur le monde iranien oriental (IXe–XVe s.), Napoli, 18–19 September
2018.
Laviola, Valentina. Islamic Metalwork from Afghanistan (9th–13th Century),
Università degli Studi di Napoli l’Orientale, Dipartimento Asia, Africa e
Mediterraneo, Series Maior 15. Rome: Unior Press, 2020.
Lawler, Andrew. “Satellites Trace Afghanistan’s Lost Empires.” Science 358(6369)
(December 15, 2017): 1364–1365.
Le Berre, Marc. “Le monument de Danestama en Afghanistan.” Révue des Études
Islamiques 38 (1970): 43–53.
Le Berre, Marc, Henri Marchal, Jean-Claude Gardin, and Bertille Lyonnet.
Monuments Pré-Islamiques de l’Hindukush Central, vol. XXIV, Mémoires de La
Délégation Archéologique Française en Afghanistan. Paris: Éditions Recherche
sur les Civilisations, 1987.

366 Iran to India
Lee, Jonathan. “Monuments of Bamiyan Province, Afghanistan.” Iran 44 (2006):

229–252.
Leshnik, Lorenz. “Ghor, Firozkoh, and the Minar-i Jam.” Central Asiatic Journal
XII(1) (1968): 36–49.
Lintz, Ulrike-Christiane. “Persisch-hebräische Inschriften aus Afghanistan.”
Judaica 64(4) (2008): 333–358.
Lintz, Ulrike-Christiane. “Persisch-hebräische Inschriften aus Afghanistan (Teil
II).” Judaica 65(1) (2009): 43–74.
Lintz, Ulrike-Christiane. “The Qur’anic Inscriptions of the Minaret of Jam
in Afghanistan,” in Calligraphy and Architecture in the Muslim World (eds.
Mohammad Gharipour and Irvin Cemil Schick), 83–102. Edinburgh:
Edinburgh University Press, 2013.
Litvinsky, Boris and Viktor Solov’ev. “The Architecture and Art of Kafyr Kala
(Early Medieval Tokharistan).” Bulletin of the Asia Institute, Aspects of Iranian
Culture, 4 (1990): 61–75.
Lohuizen-de Leeuw, J. E. van. “An Ancient Hindu Temple in Eastern
Afghanistan.” Oriental Art 5 (1959): 61–69.
Lorain, Thomas. “Chehel Dokhtaran et Shahr-e Gholgholah (Bamiyan): Nouvelles
découvertes autour de l’architecture palatiale médiévale en Afghanistan,” paper
presented at Textes et Contextes: Rechèrches en cours sur le monde iranien
oriental (IXe–XVe s.), Napoli, 18–19 September 2018.
Maitland, P. J. (Major). Records of Intelligence Party, Afghan Boundary Commission.
Diary of Major Maitland with Notes on the Population and Resources of Districts
Visited, 1884 to 1887, vol. II. Simla: Government Central Press, 1888.
Maitland, P. J. (Lieutenant-Colonel). “Chahar Aimak Tribes of Herat. No. III –
Firozkohis,” in Records of the Intelligence Party, Afghan Boundary Commission.
Reports on Tribes, Namely, Sarik Turkomans, Chahar Aimak Tribes, and
Hazaras, vol. IV, 112–141. Simla: Government Central Printing Office,
1891.
Malamud, Margaret. “The Politics of Heresy in Medieval Khurasan: The
Karramiyya in Nishapur.” Iranian Studies 27(1–4) (1994): 37–51.
Manhart, Christian. “UNESCO’s Rehabilitation of Afghanistan’s Cultural
Heritage: Mandate and Recent Activities,” in Art and Archaeology of
Afghanistan: Its Fall and Survival (ed. Juliette van Krieken-Pieters), 49–60.
Leiden: Brill, 2006.
Maricq, André and Gaston Wiet. Le minaret de Djam: La découverte de la capitale
des sultans ghorides (XIIe–XIIIe Siècles), vol. XVI: Mémoires de La Délégation
Archéologique Française en Afghanistan. Paris: Librairie C. Kinkcksieck, 1959.
Matthee, Rudi and Hiroyuki Mashita. “Kandahar Iv. From the Mongol Invasion
through the Safavid Era,” in Encyclopaedia Iranica, online edition, [2010] 2012.
Melikian-Chirvani, Assadullah Souren. “Remarques préliminaires sur un mauso-
lée ghaznévide.” Arts Asiatiques XVII (1968): 59–92.
Melikian-Chirvani, Assadullah Souren. “Le roman de Varque et Golšâh: essai
sur les rapports de l’esthétique littéraire et de l’esthétique plastique dans l’Iran
pré-mongol, suivi de la traduction du poème.” École Française d’Extrême-
Orient, 1970a.
Melikian-Chirvani, Assadullah Souren. “Eastern Iranian Architecture: Apropos of

the Ghurid Parts of the Great Mosque of Harat.” BSOAS XXXIII(2) (1970b):
322–327.
Melikian Chirvani, Assadullah Souren. Islamic Metalwork from the Islamic World,
8th–18th Centuries. London: Her Majesty’s Stationery Office, 1982.
Melikian-Chirvani, Assadullah Souren. “The Buddhist Ritual in the Literature
of Early Islamic Iran,” in South Asian Archaeology 1981, 272–279. Cambridge:
Cambridge University Press, 1984.
Melikian-Chirvani, Assadullah Souren. “Iran to Tibet,” in Islam and Tibet:
Interactions along the Musk Routes (eds. Anna Akasoy, Charles Burnett, and
Ronit Yoeli-Tlalim), 89–115. Farnham: Ashgate, 2011.
Melikian-Chirvani, Assadullah Souren. “Buddhism ii. In Islamic Times,” in
Encyclopaedia Iranica, online edition, 2013.
Moline, Judi. “The Minaret of Gam.” Kunst Des Orients 9(1/2) (1973/74): 131–148.
Morgenstierne, G. “AFGHANISTAN vi. Paṧto,” in Encyclopaedia Iranica, online
edition, 1982.
Müller-Wiener, Martina. “Relief Ware: Of Moulds and Stamps,” in Herat through
Time, vol. 3: Ancient Herat (eds. Ute Franke and Martina Müller-Wiener),
184–229. Berlin: Staatliche Museen zu Berlin, 2016.
Najimi, Abdul Wasay. “The Ghurid Madrasa and Mausoleum of Shah-i Mashhad,
Ghur, Afghanistan.” Iran LIII (2015): 143–169.
Nicol, Norman D. “A Khwarizmian–Ghorid Hoard from the Time of the
Mongol Invasion,” in Simoe Assemani Symposium on Islamic Coinage. The 2nd
International Congress on Numismatic and Monetary History, 151–190. Padua:
Esedra, 2005.
Nizami, K. A. “The Ghurids,” in History of Civilizations of Central Asia, vol. IV:
The Age of Achievement: AD 750 to the End of the Fifteenth Century (eds. M. S.
Asimov and C. E. Bosworth), 1st Indian edn., 77–190.. New Delhi: Motilal
Banarsidass Publishers, 1999.
O’Kane, Bernard. “Salguq Minarets: Some New Data.” Annales Islamologiques 20
(1984): 85–101.
O’Neal, Michael P. “The Ghurid Empire: Warfare, Kingship, and Political
Legitimacy in Eastern Iran and Northern India,” Tel Aviv University, 2013.
O’Neal, Michael P. “Mapping the Ghurid Empire: Numismatic Evidence and
Narrative Sources.” Hofstra University, Hempstead, New York, 2015.
O’Neal, Michael P. “Ghurids,” in Encyclopaedia of Islam 3rd edn. (eds. Kate Fleet,
Gudrun Krämer, Denis Matringe, John Nawas, and Everett Rowson). Leiden:
Brill online, July 21, 2016a.
O’Neal, Michael P. “The Ghazna Coinage of the Ghūrid Sultan ‘Alā’ Al-Dīn
Óusayn Jahān-Sūz.” Hofstra University, Hempstead, New York, 2016b.
O’Neal, Michael P. “The Mongol Invasion of Afghanistan,” paper presented at
the Jordan Center for Persian Studies, University of California, Irvine, 2017.
O’Neal, Michael P. “Some New Numismatic Evidence for Ghurid History,” in
Transactions of the State Hermitage Museum, 199–221. St. Petersburg: State
Hermitage Museum, 2020.
Pal, Pratapaditya. “Evidence of Jainism in Afghanistan and Kashmir in Ancient
Times.” Bulletin of the Asia Institute 21 ([2007] 2012): 25–34.
Paul, Jürgen. “The Histories of Herat.” Iranian Studies 33(1/2) (2000): 93–115.

368 Iran to India
Paul, Jürgen. “Cities in Medieval Iran: A Review of Recent Publications.” Eurasian

Studies 16 (2018a): 5–20.
Paul, Jürgen. “Balkh, from the Seljuqs to the Mongol Invasion.” Eurasian Studies
16 (2018b): 313–351.
Peacocke, Captain R. E., et al. Records of Intelligence Party, Afghan Boundary
Commission. Diary of Captain Peacocke, R.E. and Reports on Passes over the
Range North of the Herat Valley, vol. III. Simla: Government Central Press,
1887.
Pehrson, Robert N. The Social Organization of the Marri Baluch (ed. Frederik
Barth), vol. 43, Viking Fund Publications in Anthropology. New York:
Wenner-Gren Foundation for Anthropological Research, 1966.
Pinder-Wilson, Ralph. “Ghaznavid and Ghurid Minarets.” Iran 39 (2001):
155–186.
Pugachenkova, G. A. “Little Known Monuments of the Balkh Area.” Art and
Archaeology Research Papers 13 (1978): 31–40.
Rao, Aparna. Les Gorbat d’Afghanistan. Aspects économiques d’un groupe itinérant
“Jat.” Paris: Editions Rechèrche sur les Civilisations, 1982.
Rasikh, Jawan Shir. “Early Islamic Ghur, 10th–12th Centuries ce: Rereading the
Tabaqat-i Nasiri,” unpublished PhD dissertation, University of Pennsylvania,
2019.
Rasikh, Jawan Shir. “The Many Lives of a Medieval Muslim Scholar: An
Intoduction to the Life and Times of Minhaj Siraj al-Din Juzjani, 1193–1260
ce.” Afghanistan 3(2) (2020): 111–134.
Rugiadi, Martina. “A Carved Wooden Door from Jam: Preliminary Remarks.”
Iran 44 (2006): 363–365.
Rugiadi, Martina. “Documenting Marbles of the Islamic Period from the Area
of Ghazni: The Italian Contribution (1957–2007),” in The IsIAO Italian
Archaeological Mission in Afghanistan 1957–2007: Fifty Years of Research in the
Heart of Eurasia. Proceedings of the Symposium Held in the Istituto Italiano per
l’Africa e l’Oriente, Rome, 8 January 2008 (eds. Anna Filigenzi and Roberta
Giunta), 21:105–115. Conferenze. Rome: Istituto Italiano per l’Africa e l’Ori-
ente, 2009.
Rugiadi, Martina. “The Ghaznavid Marble Architectural Decoration: An
Overview,” 2010.
Scerrato, Umberto. “The First Two Excavation Campaigns at Ghazni, 1957–1958.”
East and West 10(1/2) (1959): 23–55.
Scerrato, Umberto. “Islamic Glazed Tiles with Moulded Decoration from
Ghazni.” East and West 13(2/3) (1962): 263–287.
Schlumberger, Daniel. “Le Palais Ghaznevide de Lashkari Bazar.” Syria VI(12)
(1952): 251–270.
Schlumberger, Daniel. Lashkari Bazar: une résidence royale ghaznévide et ghoride,
vol. IA: L’Architecture, 2 vols. Paris: Diffusion de Boccard, 1978.
Schneider, Madeleine. “Remarques au sujet d’une inscription du mausolée dit de
Baba Hatim.” Studia Iranica 13(1) (1984): 165–167.
Shokoohy, Mehrdad. “The Shrine of Imam-i Kalan in Sar-i Pul, Afghanistan.”
Bulletin of the School of Oriental and African Studies 52(2) (1989):
306–314.

Sing, Hira. “Appendix to Firozkohi Report. Sub-Surveyor Hira Sing’s Journey

in the Firozkohi Cuntry,” in Records of the Intelligence Party, Afghan Boundary
Commission. Reports on Tribes, Namely, Sarik Turkomans, Chahar Aimak
Tribes, and Hazaras, IV:142–204. Simla: Government Central Printing Office,
1891.
Sourdel, Dominique. Inventaire des monnaies musulmanes anciennes du Musée de
Caboul. Damascus: Institut français de Damas, 1953.
Sourdel-Thomine, Janine. “Deux minarets d’époque seljoukide en Afghanistan.”
Syria 30 (1953): 108–136.
Sourdel-Thomine, Janine. “Stèles arabes de Bust (Afghanistan).” Arabica 3(3)
(1956): 285–306.
Sourdel-Thomine, Janine. “L’art guride d’Afghanistan à propos d’un livre récent.”
Arabica 7 (1960): 273–280.
Sourdel-Thomine, Janine. “Le mausolée de Baba Hatim en Afghanistan.” Revue
des Études Islamiques 39 (1971): 290–320.
Sourdel-Thomine, Janine. Lashkari Bazar: une résidence royale ghaznévide et
ghoride, vol. IB: Le décor non-figuratif et les inscriptions, 2 vols. Paris: Diffusion
de Boccard, 1978.
Sourdel-Thomine, Janine. Le minaret ghouride de Jam, un chef d’oevre du XIIe
siècle. Paris: Diffusion de Boccard, 2004.
Stark, S. Archaeological Prospections in the Aktangi Valley System (Northern
Tadjikistan) in 2005. 2005, available at: http://www.orientarch.uni-halle.de/
sfb586/c5/2005/index.htm.
Tadjikistan) in 2006. 2006a, available at: http://www.orientarch.uni-halle.de/
sfb586/c5/2006/index.htm.
Tadjikistan) in 2006. 2006b, available at: http://www.orientarch.uni-halle.de/
sfb586/c5/2006/2006b.htm.
Stewart, Rory. The Places in Between. Orlando: Harcourt, 2004.
Szabo, Albert and Thomas J. Barfield. Afghanistan: An Atlas of Indigenous Domestic
Architecture. Austin: University of Texas Press, 1991.
Taddei, Maurizio. “Tapa Sardār. First Preliminary Report.” East and West 18(1/2)
(1968): 109–124.
Taddei, Maurizio. “A Note on the Barrow Cemetery at Kandahar,” in South
Asian Archaeology 1977: Papers from the Fourth International Conference of the
Association of South Asian Archaeologiests in Western Europe, Held in the Istituto
Universitario Orientale, Napoli (ed. Maurizio Taddei), 2:909–916. Napoli:
Istituto Universitario Orientale, 1979.
Taddei, Maurizio and Giovanni Verardi. “Tapa Sardār: Second Preliminary
Report.” East and West 28(1–4) (1978): 33–136.
Tarzi, Z. L’architecture et le décor rupestre des grottes de Bamiyan, 2 vols. Paris:
Imprimerie Nationale, 1977.
Tarzi, Z. “Les fouilles de la mission archéologique française à Bamiyan sous
la direction de Z. Tarzi,” in L’art d’Afghanistan de la préhistoire à nos jours:
nouvelles données, actes d’une journée d’étude, UNESCO, 11 mars 2005, 95–124.
CEREDAF, 2007.

370 Iran to India
Tarzi, Z. “Bamian ii. History and Monuments,” in Encyclopaedia Iranica, online

edition, 1988.
Tate, George Passman. Seistan: A Memoir on the History, Topography, Ruins, and
People of the Country. Quetta, Pakistan: Abid Bokhari, [1910] 1977.
Thomas, David. “The Ebb and Flow of Empires: Afghanistan and Neighboring
Lands in the Twelfth–Thirteenth Centuries,” in Proceedings of the 5th
International Congress on the Archaeology of the Ancient Near East, 3:285–299.
Madrid: UAM Ediciones, 2008.
Thomas, David. “Signifying the ‘Ghurid Self’: The Juxtaposition between
Historical Labels and the Expression of Identity in Material Culture.” Journal
of Historical and European Studies 3 (2010): 71–88.
Thomas, David. “The Ebb and Flow of an Empire: The Ghurid Polity of Central
Afghanistan in the Twelfth and Thirteenth Centuries,” La Trobe University,
2011.
Thomas, David. “The Metamorphosis of the Minaret of Djam: From Ghurid
‘Victory Tower’ to Symbol of the New Afghanistan and Global Cultural
Property.” Archaeological Review from Cambridge 27(2) (2012): 139–159.
Thomas, David. The Ebb and Flow of the Ghurid Empire, Adapa Monographs.
Sydney: Sydney University Press, 2018.
Thomas, David C. and Alison Gascoigne. “The Architectural Legacy of the
Seasonally Nomadic Ghurids,” in Landscapes of the Islamic World: Archaeology,
History, and Ethnography (eds. Stephen McPhillips and Paul D. Wordsworth),
169–183. Philadelphia: University of Pennsylvania Press, 2016.
Thomas, D. C., K. Deckers, M. M. Hald, M. Holmes, and M. Madella.
“Environmental Evidence from the Minaret of Jam Archaeological Project,
Afghanistan.” Iran 44 (2006): 253–276.
Thomas, David, Ustad Sayid ’Umar, Faisal Ahmad, and David Smith. “The
Rediscovery of the ‘Lost’ Minaret of Qal’a-i Zārmurgh, Sāghar, Afghanistan.”
Iran LII (2014): 133–142.
Tissot, F. “Afghanistan ix. Pre-Islamic Art,” in Encyclopaedia Iranica, online
edition, [1983] 2011, available at: http://www.iranicaonline.org/articles/
afghanistan-ix-preislamic-art.
Urban, Thomas. “Archaeological Research in Qal’a-e Ekhtyaruddin: Investigation
of the Citadel Glacis – Trench 2,” in Excavations and Explorations in Herat
City (eds. Ute Franke and Thomas Urban), Ancient Herat, 3:321–331. Berlin:
Staatliche Museen zu Berlin, 2017.
Utas, Bo. “Notes on Afghan Sufi Orders and Khanaqahs.” Afghanistan Journal
7(1) (1980): 60–67.
Verardi, Giovanni and Elio Paparatti. Buddhist Caves of Jāghūrī and Qarabāgh-e
Ghaznī, Afghanistan, Reports and Memoirs, New Series, 2. Rome: IsIAO,
2004.
Vercellin, Giorgio. “The Identification of Firuzkuh: A Conclusive Proof.” East
and West 26 (1976): 337–340.
Wannell, Bruce. “Echoes in Landscape: Western Afghanistan in 1989,” in Cairo to
Kabul (eds. Warwick Ball and Leonard Harrow), 236–247. London: Melisende,
2002.
Whitehouse, David. The Barrow Cemetery at Kandahar, vol. 36, Annali Dell’Istituto

Orientale Di Napoli, N.S. XXVI. Napoli: Istituto Orientale di Napoli,

1976.
Whitehouse, David. “Sites A and B: The Barrow Cemetery,” in Excavations at
Kandahar 1974 and 1975 (eds. Anthony McNicoll and Warwick Ball), 1–18.
Oxford: Tempus Reparatum, 1996.
Wordsworth, Paul. “The Hydrological Networks of the Balkh Oasis after the
Arrival of Islam: A Landscape Archaeological Perspective.” Afghanistan 1(1)
(2018): 182–208.
Yate, C. E. “Notes on the City of Hirat.” Journal of the Asiatic Society of Bengal 56
(1887): 84–106.
Works on Pakistan
Ahmad, Shamsuddin. A Guide to Tattah and the Makli Hill. Karachi: Manager of
Publications, Government of Pakistan Press, 1952.
Ali, Taj. Anonymous Tombs in the Gomal Valley and the Beginning of Tomb
Architecture in Pakistan, Memoirs of the Department of Archaeology,
University of Peshawar No. 4. Peshawar: Department of Archaeology,
University of Peshawar, 1988.
Ali, Taj. “Medieval Architectural Remains near Kabirwala, Khanewal District.”
Ancient Pakistan 8 (1993a): 125–131.
Ali, Taj. “Tomb of Shaikh Sadan Shahid, Its Decoration.” Ancient Pakistan 8
(1993b): 133–139.
Andrews, Peter Alford. Felt Tents and Pavilions: The Nomadic Tradition and Its
Interaction with Princely Tentage, 2 vols. London: Melisende, 1999.
Azeem, Rizwan. “Evolution of Sultanate Period Architecture in Multan,” in
Sultanate Period Architecture in Pakistan. Proceedings of the Seminar Held in
Lahore, November 1990 (eds. Siddiq Akbar, Abdul Rehman, and Muhammad
Ali Tirmizi), Architectural Heritage of Pakistan II, 87–100. Lahore: Anjuman
Mimaran, 1991.
Bagnera, Alessandra. “Preliminary Note on the Islamic Settlement of Udegram,
Swat: The Islamic Graveyard (11th–13th Century A.D.).” East and West 56(1/3)
(2006): 205–228.
Baloch, N. A. “The Kalmati Tombs in Sindh and Balochistan.” Pakistan
Archaeology 26 (1991): 243–256.
Bokhari, Hakim Ali Shah. “Chaukhandi Type Stone Tombs at Taung, History
and Conservation.” Pakistan Archaeology 27 (1992): 89–99.
Cousens, Henry. The Antiquities of Sind with Historical Outline, vol. 46,
Archaeological Survey of India, Imperial Series. Calcutta: Government of
India Central Publication Branch, 1929.
Dani, A. H., H. Humbach, and R. Gobl. “Tochi Valley Inscriptions in the
Peshawar Museum.” Ancient Pakistan 1 (1964): 125–135.
Dani, A. H. Thatta: Islamic Architecture. Islamabad: Institute of Islamic History,
Culture and Civilization, 1982.
Edwards, Holly. “The Genesis of Islamic Architecture in the Indus Valley,” PhD
dissertation, New York University, 1990.

372 Iran to India
Edwards, Holly. “The Ribat of ’Ali B. Karmakh.” Iran 29 (1991): 85–94.

Edwards, Holly. “Centralizing the Margins: Commemorative Architecture in the
Indus Valley,” in The Architecture of the Indian Sultanates (eds. Abha Narain
Lambah and Alka Patel), 18–29. Mumbai: Marg Publications, 2006.
Edwards, Holly. Of Brick and Myth: The Genesis of Islamic Architecture in the Indus
Valley. Karachi: Oxford University Press, 2015.
Farooq, Abdul Aziz. “Mosque or Khalid Walid’s Tomb (A Ghorid Monument,
Khanewal District),” n.d.
Flood, Finbarr B. “Ghurid Architecture in the Indus Valley: The Tomb of Shaykh
Sadan Shahid.” Ars Orientals 31 (2001): 129–166.
Friedmann, Y. and P. A. Andrews. “Multān,” in Encyclopaedia of Islam, 2nd edn
(eds. P. Bearman, Th. Bianquis, C. E. Bosworth, E. van Donzel, and W. P.
Heinrichs). Leiden: Brill online, 2012.
Ghafur, M. A. “Two Lost Inscriptions Relation to the Arab Conquest of Kabul
and the North West Region of West Pakistan.” Ancient Pakistan 2 (1965/66):
4–12.
Haig, Malcolm Robert. The Indus Delta Country: A Memoir, Chiefly on Its Ancient
Geography and History, with Three Maps. London and Karachi: Kegan Paul,
Trench, Trubner and Indus Publications, 1894.
Hasan, Sheikh Khursid. “Pre-Mughal Tombs at Alore District Sukkur (Sindh).”
Journal of Central Asia XIV(1) (1991): 123–129.
Hasan, Sheikh Khursid. Chaukhandi Tombs in Pakistan. Karachi: Royal Book
Company, 1996.
Hassan, M. Usman (Brig. retd.). “Terracotta Plaque Decorated Tombs of
Baluchistan,” in Sultanate Period Architecture in Pakistan. Proceedings of the
Seminar Held in Lahore, November 1990 (eds. Siddiq Akbar, Abdul Rehman,
and Muhammad Ali Tirmizi), Architectural Heritage of Pakistan II, 78–86.
Lahore: Anjuman Mimaran, 1991.
Ibrahim, Asma and Kaleem Lashari. “Recent Archaeological Discoveries in the
Lower Deltaic Area of Indus.” Journal of Pakistan Archaeologists Forum 2(1/2)
(1993): 1–44.
Kalhoro, Zulfiqar Ali. Perspectives on the Art and Architecture of Sindh. Karachi:
Endowment Fund Trust for Preservation of the Heritage of Sindh, 2014.
Kevran, Monik. “The Fortress of Ratto Kot, at the Mouth of the Banbhore River
(Indus Delta–Sindh–Pakistan.” Pakistan Archaeology 27 (1992): 143–170.
Kevran, Monik. “Vanishing Medieval Cities of the Northwest Indus Delta.”
Pakistan Archaeology 28 (1993): 3–54.
Kevran, Monik. “Le Port Multiple des Bouches de l’Indus: Barbariké, Deb,
Daybul, Lahori Bandar, Diul Sinde,” in Sites et Monuments Disparus D’après les
Témoignages de Voyageurs (ed. Rika Gyselen), 45–92, Res Orientales, 8. Bures-
sur-Yvette: Groupe pour l’étude de la civilisation du Moyen-Orient, 1996a.
Kevran, Monik. “Entre l’Inde et l’Asie centrale: Les mausolées islamique du Sind
et du Penjab.” Cahiers D’Asie Centrale 1/2 (1996b): 133–171.
Kevran, Monik. “Caravanserails du delta de l’Indus: Réflexions sur l’origine du
caravanserail islamique.” Archéologie Islamique 8/9 (1999a): 143–176.
Kevran, Monik. “Multiple Ports at the Mouth of the River Indus: Barbarike,
Deb, Daybul, Lahori Bandar, Diul Sinde,” in The Archeology of Seafaring:

The Indian Ocean in the Ancient Period (ed. Himanshu Prabha Ray), 70–154.
Delhi: Pragati Publications, 1999b.
Khan, Ahmad Nabi. Multan: History and Architecture. Islamabad: Institute of
Islamic History, Culture & Civilization, Islamic University, 1983.
Khan, Ahmad Nabi. “A Group of Four Tombs in the Multan Style of Architecture
at Lal Muhra Sharif (D. I. Khan).” Journal of Central Asia 7(1) (1984): 29–48.
Khan, Ahmad Nabi. “Naked Brick Architecture of Early Islamic Period of
Pakistan.” Pakistan Archaeology 23 (1988): 303–325.
Khan, Ahmad Nabi. Islamic Architecture in South Asia, Pakistan–India–Bangladesh.
Oxford: Oxford University Press, 2003.
Khan, F. A. Banbhore: A Preliminary Report on the Recent Archaeological Excavations
at Banbhore, 2nd edn. Pakistan: Department of Archaeology and Museums,
1963.
Khan, Gulzar Muhammad. “Pre-Mughal Mosques in Lahore,” in Sultanate Period
Architecture in Pakistan. Proceedings of the Seminar Held in Lahore, November
1990 (eds. Siddiq Akbar, Abdul Rehman, and Muhammad Ali Tirmizi),
Architectural Heritage of Pakistan II, 126–133. Lahore: Anjuman Mimaran,
1991.
Khan, Hasan Ali. Constructing Islam on the Indus: The Material History of the
Suhrawardi Sufi Order, 1200–1500 AD. New York: Royal Asiatic Society of
Great Britain and Ireland, Cambridge University Press, 2016.
Manna, Gabriella. “Some Observations on the Pottery from the Islamic Settlement
of Udegram, Swat.” East and West 56(1/3) (2006): 229–236.
Masson, Charles. Narrative of a Journey to Kalât, Including an Account of the
Insurrection at That Place in 1840; and a Memoir on Eastern Balochistan.
London: Richard Bentley, 1843.
Meister, Michael. “Temples along the Indus.” Expedition 38(3) (1996): 41–54.
Meister, Michael. “Gandhara-Nagara Temples of the Salt Range and Indus.”
Kala, The Journal of the Indian Art History Congress 4 (1997/98): 45–52.
Meister, Michael. “Temples of the Salt Range,” in Religion, Ritual, and Royalty
(eds. N. R. Singhi and R. Joshi), 132–139. Jaipur: Rawat Publication, 1999.
Meister, Michael. “Pattan Munara : Minar or Mandir ?” in Hari Smriti Studies
on Art Archaeology and Indology (ed. Banerji Arundhati), 113–121. New Delhi-
110002: Kaveri Books, 2006.
Meister, Michael. “Exploring Kafirkot: When Is a Rose Apple Not a Rose.”
Pakistan Heritage 1 (2009): 109–128.
Meister, Michael. Temples of the Indus: Studies in the Hindu Architecture of Ancient
Pakistan, Brill’s Indological Library, vol. 35. Leiden: Brill, 2010.
Meister, Michael. “Continuities of Architectural Heritage in the Northwest,” paper
presented at American Council of Southern Asian Art Symposium XV,
Minneapolis, MN, 2011.
Meister, Michael, Abdur Rehman, and Farid Khan. “Discovery of a New Temple
on the Indus.” Expedition 42(1) (2000): 37–46.
Mumtaz, Kamil Khan. Architecture in Pakistan. Singapore: Concept Media, 1985.
Pottinger, Henry. Travels in Beloochistan and Sinde; Accompanied by a Geographical
and Historical Account of Those Countries. London: Longman, Hurst, Rees,
Orme & Brown, 1816.

374 Iran to India
Rahman, Abdur. The Last Two Dynasties of the Śahis: An Analysis of Their History,
Archaeology, Coinage, and Palaeography. Islamabad: Centre for the Study of the
Civilizations of Central Asia, Quaid-i-Azam University, 1979.
Rahman, Abdur. “The Zalamkot Bilingual Inscription.” East and West 48(3/4)
(1998): 469–473.
Rahman, Abdur. “Arslan Jadhib, Governor of Tus: The First Muslim Conqueror
of Swat.” Ancient Pakistan 15 (2002a): 11–14.
Rahman, Abdur. “New Light on the Khingal, Turk and the Hindu Sahis.” Ancient
Pakistan 15 (2002b): 37–42.
Rehman, Abdul. “Sultanate Period Architecture in the Punjab (1000 A.D.–
1500 A.D.),” in Sultanate Period Architecture in Pakistan. Proceedings of the
Seminar Held in Lahore, November 1990 (eds. Siddiq Akbar, Abdul Rehman,
and Muhammad Ali Tirmizi), Architectural Heritage of Pakistan II, 36–58.
Lahore: Anjuman Mimaran, 1991.
Rehman, Abdul and Talib Hussain. “Expression of Paying Tribute to the Saint:
Decorative Vocabulary on the Tomb of Ahmad Kabir.” Journal of Research in
Architecture and Planning 10(1) (2011): 59–75.
Scerrato, Umberto. “Research on the Archaeology and History of Islamic Art in
Pakistan: Excavation of the Ghaznavid Mosque on Mt. Raja Gira, Swat.” East
and West, 35 (1985): 439–451.
Sharma, Sunil. Persian Poetry at the Indian Frontier: Mas‘ud Sa‘d Salman of Lahore,
Permanent Black Monographs, Opus 1 Series. New Delhi: Permanent Black,
2000.
Stein, Aurel and Sten Konow. An Archaeological Tour in Waziristān and Northern
Balūchistān, Memoirs of the Archaeological Survey of India, No. 37. Calcutta:
Government of India Central Publication Office, 1929.
Stein, Aurel and Sten Konow. An Archæological Tour in Upper Swāt and Adjacent
Hill Tracts, Memoirs of the Archaeological Survey of India, No. 42. Calcutta:
Government of India Central Publication Branch, 1931.
Vogel, Jean Philippe. “Tombs at Hinidan in las Bela.” Annual Report (1902–1903):
213–217.
Zajadacz-Hastenrath, Salome. Chaukhandigr.ber: Studien sur Grabkunst in Sind u.
Baluchistan. Wiesbaden: Steiner, 1978.
Ziad, Waleed. “‘Islamic Coins’ from a Hindu Temple: Reconsidering Ghaznavid
Interactions with Hindu Sacred Sites through New Numismatic Evidence
from Gandhara.” Journal of the Economic and Social History of the Orient 59
(2016): 618–659.
Anthropological Works
Amiri, Ghulam Rahman. The Helmand Baluch: A Native Ethnography of the

People of Southwest Afghanistan, ed. William B. Trousdale, trans. James
Gehlhar, Mhairi Gehlhar, and Babrak Amiri. New York: Berghahn Books,
2020.
Barfield, Thomas. The Central Asian Arabs of Afghanistan: Pastoral Nomadism in
Transition. Austin: University of Texas Press, 1981.

Barfield, Thomas. The Nomadic Alternative. Englewood Cliffs, NJ: Prentice Hall,
1993.
Barfield, Thomas. “Conclusion,” in Nomadic Pathways in Social Evolution (eds.
N. N. Kradin and D. M. Bondarenko), English edn., 172–179, “Civilizational
Dimension” Series. Moscow/Lac-Beauport: MEA Books, 2003.
Barth, Frederik. Nomads of South Persia: The Basseri Tribe of the Khamseh
Confederacy, 2nd edn. Oslo and Boston: Oslo University Press and Little,
Brown, 1961.
Beck, Lois. “Chapter 9: Iran and the Qashqai Confederacy,” in The Conflict
of Tribe and State in Iran and Afghanistan (ed. Richard Tapper), 284–313.
London: Croom Helm, 1983.
Bronson, Bennet. “The Role of Barbarians in the Fall of States,” in The Collapse of
Ancient States and Civilizations (eds. Norman Yoffee and George L. Cowgill),
196–218. Tucson: University of Arizona Press, 1988.
Büssow, Johann, David Durand-Guédy, and Jürgen Paul (eds.). “Nomads in the
Political Field.” Eurasian Studies IX (2011): 1–9.
Cribb, Roger. Nomads in Archaeology. Cambridge: Cambridge University Press,
1991.
Elliott, Chris. “Understanding the Dialectic of Nomad and State.” Small Wars
Journal, December 2, 2013, available at: http://smallwarsjournal.com/jrnl/art/
understanding-the-dialectic-of-nomad-and-state-0#_ednrefl.
Garthwaite, Gene. “Tribes, Confederation, and the State: An Historical
Overview of the Bakhtiari and Iran,” in The Conflict of Tribe and State in
Iran and Afghanistan (ed. Richard Tapper), 314–336. London: Croom Helm,
1983.
Gellner, Ernest. “The Tribal Society and Its Enemies,” in The Conflict of Tribe and
State in Iran and Afghanistan (ed. Richard Tapper), 436–448. London: Croom
Helm, 1983.
Glatzer, Bernt. “Political Organisation of Pashtun Nomads and the State,” in
The Conflict of Tribe and State in Iran and Afghanistan (ed. Richard Tapper),
212–232. London: Croom Helm, 1983.
Glatzer, Bernt and Michael Casimir. “Herds and Households among Pashtum
Pastoral Nomads: Limits of Growth.” Ethnology 22(4) (1983): 307–327.
Irons, William. “Cultural Capital, Livestock Raiding, and the Military Advantage
of Traditional Pastoralists,” in Nomadic Pathways in Social Evolution (eds.
Nikolay N. Kradin, D. M. Bondarenko, and Thomas J. Barfield), English
edn., 5:73–87, “Civilizational Dimension” Series. Moscow/Lac-Beauport:
MEA Books, 2003.
Kradin, Nikolay N. “Nomadic Empires: Origins, Rise, Decline,” in Nomadic
Pathways in Social Evolution (eds. Nikolay N. Kradin, D. M. Bondarenko, and
Thomas J. Barfield), English edn, 73–87, “Civilizational Dimension” Series.
Moscow/Lac-Beauport: MEA Books, 2003.
Næss, Marius Warg. “Predatory or Prey – the Rise of Nomadic Empires.”
Pastoralism, Climate Change and Policy (blog), November 20, 2015, available at:
https://pastoralism-climate-change-policy.com/2015/11/20/predatory-or-prey-
the-rise-of-nomadic-empires.
Pijl, Kees van der. Nomads, Empires, States. London: Pluto Press, 2007.

376 Iran to India
Tapper, Richard (ed.). The Conflict of Tribe and State in Iran and Afghanistan.
London and New York: Croom Helm and St Martin’s Press, 1983.
Tapper, Richard. Frontier Nomads of Iran: A Political and Social History of the
Shahsevan, Cambridge Middle East Studies, 7. Cambridge: Cambridge
University Press, 1997.
Vasjutin, Sergey A. “Typology of Pre-States and Statehood Systems of Nomads,”
in Nomadic Pathways in Social Evolution (eds. N. N. Kradin and D. M.
Bondarenko), 50–62, English edn, “Civilizational Dimension” Series. Moscow/
Lac-Beauport: MEA Books, 2003.
Other Works
Akhmedov, A. “Chapter 7, Astronomy, Astrology, Observations and Calendars,”

in History of Civilizations of Central Asia, vol. IV: The Age of Achievement: AD
750 to the End of the Fifteenth Century – Part Two: The Achievements (eds. C. E.
Bosworth and M. S. Asimov), 195–204. Paris: UNESCO Publishing, 2000.
Alam, Muzaffar. “The Culture and Politics of Persian in Precolonial Hindustan,”
in Literary Cultures in History: Reconstructions from South Asia (ed. Sheldon
Pollock), 131–198. Berkeley: University of California Press, 2003.
Alam, Muzaffar. The Languages of Political Islam in India, C.1200–1800. Delhi:
Permanent Black, 2004.
Ali, Daud. Courtly Culture and Political Life in Early Medieval India. Cambridge
Studies in Indian History and Society, 10. Cambridge: Cambridge University
Press, 2004.
Allegranzi, Viola. “Vers un réexamen des inscriptions historiques du monde
iranien pré-mongol: Étude des cas des mausolées de Tim et de Termez en
Ouzbékistan,” in Texts and Contexts: Ongoing Researches on the Eastern Iranian
World (Ninth–Fifteenth C.) (eds. Viola Allegranzi and Valentina Laviola),
103–134. Rome: Istituto per l’Oriente C. A. Nallino, 2020.
Amin, Shahid. “On Retelling the Muslim Conquest of North India,” in History
and the Present (eds. Partha Chatterjee and Anjan Ghosh), 24–43. New Delhi:
Amin, Shahid. Conquest and Community: The Afterlife of Warrior Saint Ghazi
Miyan. Chicago: University of Chicago Press, 2015.
Amitai, R. “Toward a Pre-History of the Islamization of the Turks: A Re-Reading
of Ibn Fadlan’s Rihla,” in Islamisation de l’Asia Centrale (ed. Étienne de la
Vassière), 277–296. Paris: Association pour l’avancement des études irani-
ennes, 2008.
Anawati, Claude. “Fakhr Al-Dīn al-Rāzī,” in Encyclopedia of Islam, 2nd edn.
Anjum, Tanvir. Chishti Sufis in the Sultanate of Delhi 1190–1400. Karachi: Oxford
Anooshahr, Ali. “‘Utbi and the Ghaznavids at the Foot of the Mountain.” Iranian
Studies 38(2) (June 2005): 271–291.
Anooshahr, Ali. “Mughal Historians and the Memory of the Islamic Conquest

of India.” Indian Economic and Social History Review 43(3) (2006): 275–300.
Anooshahr, Ali. The Ghazi Sultans and the Frontiers of Islam. Abingdon: Routledge,
2009.
Anooshahr, Ali. “The Shaykh and the Shah: On the Five Jewels of Muhammad
Ghaws Gwaliori,” in India and Iran in the Longue Durée (eds. Alka Patel and
Touraj Daryaee), 91–102. Irvine, CA: Jordan Center for Persian Studies, 2017.
Anooshahr, Ali. “The Elephant and the Sovereign: India circa 1000 ce.” Journal of
the Royal Asiatic Society 28(4) (2018a): 615–644.
Anooshahr, Ali. Turkestan and the Rise of Eurasian Empires: A Study of Politics and
Invented Traditions. Oxford: Oxford University Press, 2018b.
Ansari, A. S., Bazmee. “Gakkhaŕ,” in Encyclopaedia of Islam, 2nd edn (eds.
P. Bearman, Th. Bianquis, C. E. Bosworth, and W. P. Heinrichs). Leiden:
Brill online, 2012.
Aquil, Raziuddin. Sufism, Culture, and Politics: Afghans and Islam in Medieval
North India. New Delhi: Oxford University Press, 2007.
Asher, Catherine. Delhi’s Qutb Complex: The Minar, Mosque and Mehrauli.
Mumbai: Marg Publications, 2017.
Asif, Manan Ahmed. A Book of Conquest: The Chachnama and Muslim Origins in
South Asia. Cambridge, MA: Harvard University Press, 2016.
Asif, Manan Ahmed. The Loss of Hindustan: The Invention of India. Cambridge,
MA: Harvard University Press, 2020.
Auer, Blain H. Symbols of Authority in Medieval Islam: History, Religion and
Muslim Legitimacy in the Delhi Sultanate, Library of South Asian History and
Culture, 6. London: I. B. Tauris, 2012.
Auer, Blain H. “Chishtiyya,” in Encyclopaedia of Islam, 3rd edn. (eds. Kate Fleet,
Gudrun Krämer, Denis Matringe, John Nawas, and Everett Rowson), Leiden:
Brill online, 2016.
Auer, Blain H. “Chapter 4. Persian Historiography in India,” in Literature from
Outside Iran: The Indian Subcontinent, Anatolia, Central Asia, and in Judeo-
Persian (ed. John Perry), 94–139. London: I. B. Tauris, 2018.
Azarpay, Guitty. “The Islamic Tomb Tower: A Note on Its Genesis and
Significance,” in Essays in Islamic Art and Architecture: In Honor of Katharina
Otto-Dorn (eds. Abbas Daneshvari and Katharina Otto-Dorn), 9–11. Islamic
Art and Architecture, 1. Malibu, CA: Undena Publications, 1981.
Balasubramaniam, R. The World Heritage Complex of the Qutub. New Delhi:
Aryan Books, 2005.
Barthold, W. Turkestan down to the Mongol Invasion, 3rd edn., vol. V, E. J. W.
Gibb Memorial Series. London: Luzac, 1968.
Berggren, J. L. “Chapter 6, Mathematical Sciences; Part Two, the Mathematical
Sciences,” in History of Civilizations of Central Asia, vol. IV: The Age of
Achievement: AD 750 to the End of the Fifteenth Century – Part Two: The
Achievements (eds. C. E. Bosworth and M. S. Asimov), 182–193.. Paris:
UNESCO Publishing, 2000.
Bhandare, Shailendra. “Transregional Connections: The ‘Lion and Sun’ Motif
and Coinage between Anatolia and India,” in Turkish History and Culture in
India (eds. A. C. S. Peacock and Richard P. McClary), 203–247. Leiden: Brill,
2020.

378 Iran to India
Bhandarkar, D. R. “The Chahamanas of Marwar.” Epigraphia Indica XI (1911/12):

26–79.
Biran, Michal. “True to Their Ways: Why the Qara Khitai Did Not Convert to
Islam,” in Eurasian Nomads and the Sedentary World (eds. Reuven Amitai and
Michal Biran), 175–199. Leiden: E. J. Brill, 2004.
Biran, Michal. The Empire of the Qara Khitai in Eurasian History: Between China
and the Islamic World, Cambridge Studies in Islamic Civilization. Cambridge:
Biran, Michal. “Jovayni, Íāªeb divan,” in Encyclopaedia Iranica, online edition,
[2009] 2012.
Blair, Sheila S. “The Madrasa at Zuzan: Islamic Architecture in Eastern Iran on
the Eve of the Mongol Invasion.” Muqarnas 3 (1985): 76–91.
Blair, Sheila S. The Monumental Inscriptions from Early Islamic Iran and
Transoxiana. Leiden: Brill, 1992.
Blair, Sheila S. Islamic Inscriptions. Edinburgh: Edinburgh University Press, 1998.
Blayac, Johanna. “Sovereign Epigraphy in Location: Politics, Devotion and
Legitimisation around the Qutb Minar, Delhi,” in Calligraphy and Architecture
in the Muslim World (eds. Mohammad Gharipour and Irvin Cemil Schick),
217–229. Edinburgh: Edinburgh University Press, 2013.
Bloom, Jonathan M. Minaret: Symbol of Islam, Oxford Studies in Islamic Art, vol.
VII. Oxford: Oxford University Press, 1989.
Bopearachchi, Osmund. The Pleasure Gardens of Sigiriya: A New Approach.
Colombo: Godage Book Emporium, 2006.
Bosworth, C.E. “The Political and Dynastic History of the Iranian World (AD
1000–1217),” in The Cambridge history of Iran. Volume V: the Saljuq and
Mongol Periods.(ed. J.A. Boyle), 1–202. Cambridge: Cambridge University
Press, 1968b.
Bosworth, Clifford Edmund. “The Heritage of Rulership in Early Islamic Iran
and the Search for Dynastic Connections with the Past.” Iran 11 (1973): 51–62.
Bosworth, C. E. and M. S. Asimov (eds.). History of Civilizations of Central Asia,
vol. IV: The Age of Achievement: AD 750 to the End of the Fifteenth Century
– Part Two: The Achievements, Indian edn. New Delhi: Motilal Banarsidass
Publishers, 2000.
Boucharlat, Rémy. “Découvertes récentes de stucs en Iran sassanide et problèmes
d’interprétation.” Syria Supp. V (2019): 351–366.
Böwering, Gerhard. “ČEŠTĪYA,” in Encyclopaedia Iranica, online edition, [1991]
2011.
Böwering, Gerhard and Matthew Melvin-Khoushki. “Khanaqah,” in Encyclopaedia
Iranica, online edition, [2010] 2012.
Bühler, G. “Eleven Land-Grants of the Chaulukyas of Anhilvad: A Contribution
to the History of Gujarat.” Indian Antiquary 6 (1877): 180–214.
Bulliet, Richard W. The Patricians of Nishapur. Cambridge, MA: Harvard
Bulliet, Richard W. “Local Politics in Eastern Iran under the Ghaznavids and
Seljuks.” Iranian Studies 11(1/4) (1978): 35–56.
Bulliet, Richard W. Islam: The View from the Edge. New York: Columbia

Burbank, Jane and Frederick Cooper. Empires in World History: Power and the
Politics of Difference. Princeton, NJ: Princeton University Press, 2010.
Buswell, Robert and Donald Lopez. The Princeton Dictionary of Buddhism.
Princeton, NJ: Princeton University Press, 2014.
Cahen, Claude, G. Deverdun, and P. M. Holt. “Ghuzz,” in Encyclopedia of Islam,
2nd edn (eds. P. Bearman, Th. Bianquis, C. E. Bosworth, E. van Donzel, and
W. P. Heinrichs), Leiden: Brill online, 2012.
Canepa, Mathew P. “Building a New Vision of the Past in the Sasanian Empire:
The Sanctuaries of Kayansih and the Great Fires of Iran.” Journal of Persianate
Studies 6 (2013): 64–90.
Canepa, Mathew P. “Dynastic Sanctuaries and the Transformation of Iranian
Kingship between Alexander and Islam,” in Persian Kingship and Architecture:
Strategies of Power in Iran from the Achaemenids to the Pahlavis (eds. Sussan
Babaie and Talinn Grigor), 65–118. London: I. B. Tauris, 2015.
Chatterjee, Kumkum. “Scribal Elites in Sultanate and Mughal Bengal.” The
Indian Economic and Social History Review 47(4) (2010): 445–472.
Chattopadhyaya, B. D. Representing the Other? Sanskrit Sources and the Muslims.
Delhi: Manohar, 1998.
Chattopadhyaya, B. D. The Making of Early Medieval India. Delhi: Oxford
Collinet, Annabelle. “Nouvelles rechèrches sur la céramique de Nishapur:
la prospection du shahrestan,” in Greater Khorasan: History, Geography,
Archaeology and Material Culture (ed. Rocco Rante), 89–114. Berlin: De
Gruyter, 2015.
Crossley, Pamela Kyle. Hammer and Anvil: Nomad Rulers at the Forge of the
Modern World. Lanham, MD: Rowman & Littlefield, 2019.
Dähne, Burkart. “Recent Researches of the Uighur Capital Kharabalgasun (Ördu
Balik) with Focus on the So-Called Temple – or Palace Area.” Drevnie Kul’tury
Mongolii, Baı̆kal’skoı̆ Sibiri i Severnogo Kitai︠︡ a: Materialy VII Mezhdunarodnoı̆
Nauchnoı̆ Konferent︠︡ sii 4 (2013): 21–28.
Dale, Stephen F. Indian Merchants and Eurasian Trade, 1600–1750, Cambridge
Studies in Islamic Civilization. Cambridge: Cambridge University Press, 1994.
Dallapiccola, Anna Libera. “Vahanas,” in Brill’s Encyclopedia of Hinduism (eds.
Knut A. Jacobsen, Helene Basu, Angelika Malinar, and Vasudha Narayanan).
de Bruijn, J. T. P. Of Piety and Poetry: The Interaction of Religion and Literature
in the Life and Works of Hakīm Sanāʼī of Ghazna, publication of the de Goeje
Fund, No. 25. Leiden: Brill, 1983.
de Bruijn, J. T. P. “Sanā’i,” in Encyclopaedia Iranica, online edition, 2012.
de Laugier de Beaureceuil, S. “‘Abdallah Ansari,” in Encyclopaedia Iranica, online
edition, [1982] 2011.
Dani, Ahmad Hasan. “Part Two: Southern Central Asia,” in History of Civilizations
of Central Asia, vol. IV: The Age of Achievement: AD 750 to the End of the
Fifteenth Century – Part Two: The Achievements (eds. C. E. Bosworth and M. S.
Asimov), 557–584. Paris: UNESCO Publishing, 2000.
Daryaee, Touraj. “The Persian Gulf Trade in Late Antiquity.” Journal of World
History 14(1) (2003): 1–16.

380 Iran to India
Deloche, Jean. Transport and Communications in India Prior to Steam Locomotion,

2 vols, trans. James Walker, French Studies in South Asian Culture and
Society, 7. Delhi: Oxford University Press, [1980] 1993/94.
Desai, Madhuri. “Interpreting an Architectural Past: Ram Raz and the Treatise
in South Asia.” Journal of the Society of Architectural Historians 72(4) (2012):
462–487.
Desai, Z. A. “Cultural Relations between Afghanistan and India during the
Medieval Period.” Afghanistan 28(3) (1975): 9–17.
Dhaky, M. A. “Kiradu and the Maru–Gurjara Style of Temple Architecture.”
Bulletin of the American Academy of Benares 1 (1967): 35–45.
Dhaky, M. A. “Maru-Gurjara Vastu-Sastraman Masjid-Nirmana-Vidhi.”
Swadhyaya 7(1) (1968): 64–79 (in Gujarati).
Dhaky, M. A. “The Genesis and Development of Maru-Gurjara Temple
Architecture,” in Studies in Indian Temple Architecture (ed.Pramod Chandra),
114–165. Varanasi: American Institute of Indian Studies, 1975.
Digby, Simon. “Early Pilgrimages to the Graves of Mu’in al-Din Sijzi and Other
Indian Chishti Shaykhs,” in Islamic Society and Culture: Essays in Honour of
Professor Aziz Ahmad (eds. Milton Israel and N. K. Wagle), 95–100. New
Delhi: Manohar, 1983.
Dodd, Erica Cruikshank, and Shereen Khairallah. The Image of the Word: A Study
of Quranic Verses in Islamic Architecture, 2 vols. Beirut: American University
of Beirut, 1981.
Durand-Guédy, David. Iranian Elites and Turkish Rulers: A History of Isfahan
in the Saljuq Period, Routledge Studies in the History of Iran and Turkey.
London: Routledge, 2010.
Durand-Guédy, David. “The Tents of the Saljuqs,” in Turko-Mongol Rulers, Cities
and City Life (ed. David Durand-Guédy), 149–189. Leiden: Brill, 2013a.
Durand-Guédy, David. “Ruling from the Outside: A New Perspective on
Turkish Kingship in Iran,” in Every Inch a King: Comparative Studies on Kings
and Kingship in the Ancient and Medieval Worlds (eds. Lynette Mitchell and
Charles Melville), 325–342. Leiden: Brill, 2013b.
Durand-Guédy, David. “Pre-Mongol Khurasan. A Historical Introduction,” in
Greater Khorasan: History, Geography, Archaeology and Material Culture (ed.
Rocco Rante), 1–8. Berlin: De Gruyter, 2015.
Durand-Guédy, David. “Khargah and Other Terms for Tents in Firdawsi’s Shah-
Namah.” Iranian Studies 51(6) (2018): 819–849.
Edwards, Holly. “Text, Context, Architext: The Qur’an as Architectural
Inscription,” in Brocade of the Pen: The Art of Isalmic Writing (ed. Carol Garret
Fisher), 63–75. East Lansing, MI: Kresge Art Museum, 1991.
Ernst, Carl W. Eternal Garden: Mysticism, History, and Politics at a South Asian
Sufi Center. Albany, NY: State University of New York Press, 1992.
Ernst, Carl W. and Bruce Lawrence. Sufi Martyrs of Love: The Chishti Order in
South Asia and Beyond. New York: Palgrave Macmillan, 2002.
Ettinghausen, Richard. “Arabic Epigraphy: Communication or Symbolic
Affirmation,” in Near Eastern Numismatics, Iconography, Epigraphy, and History
Studies in Honor of George C. Miles, 297–317. Beirut: American University of
Beirut, 1974.

Falk, Harry. Asokan Sites and Artefacts: A Source-Book with Bibliography. Mainz
am Rhein: Verlag Philipp von Zabern, 2006.
Ferrier, J. P. Caravan Journeys and Wanderings in Persia, Afghanistan, Turkistan,
and Beloochistan; with Historical Notices of the Countries Lying between Russia
and India. trans. Capt. William Jesse. Karachi: Oxford University Press,
1976.
Finster, Barbara. “The Saljuqs as Patrons,” in The Art of the Saljuqs in Iran
and Anatolia (ed. Robert Hillenbrand), 17–23. Costa Mesa, CA: Mazda,
1994.
Flood, Finbarr B. “Between Ghazna and Delhi: Lahore and Its Lost Manara,” in
Cairo to Kabul (eds. Warwick Ball and Leonard Harrow), 102–112. London:
Melisende, 2002.
Flood, Finbarr B. “Refiguring Iconoclasm in the Early Indian Mosque,” in
Negating the Image (eds. Anne McClanan and Jeff Johnson), Burlington, VT:
Ashgate, 2005.
Flood, Finbarr B. Objects of Translation: Material Culture and Medieval “Hindu–
Muslim” Encounter. Princeton, NJ: Princeton University Press, 2009a.
Genito, Bruno. “Chapter 10. Landscape, Sources and Architecture at the
Archaeological Remains of Achaemenid Sistān (East Irān): Dāhān-i
Ghūlāmān,” in Excavating an Empire: Achaemenid Persia in Longue Durée
(eds. Touraj Daryaee, Ali Mousavi, and Khodadad Rezakhani), 163–178. Costa
Mesa, CA: Mazda, 2014.
Ghanimati, Soroor. “New Perspectives on the Chronological and Functional
Horizons of Kuh-e Khwaja in Sistan.” Iran 38 (2000): 137–150.
Godard, André. “Khorasan.” Athar-i Iran 4 (1949): 7–150.
Godard, André. “L’origine de la madrasa, de la mosquée et du caravansérail à
quatre īwāns.” Ars Islamica 15/16 (1951): 1–9.
Godard, Y. A. “Notice Épigraphique.” Athar-i Iran I(2) (1936): 361–373.
Godard, Y. A. and Myron Bement Smith. “Material for a Corpus of Early Islamic
Architecture: II. Manar and Masdjid, Barsian (Isfahan).” Ars Islamica 4 (1937):
6–41.
Golden, Peter. “Chapter One: Courts and Court Culture in the Proto-Urban and
Urban Developments among the Pre-Chinggisid Turkic Peoples,” in Turko-
Mongol Rulers, Cities and City Life (ed. David Durand-Guédy), 21–73. Leiden:
Brill, 2013.
Grabar, Oleg. “The Earliest Islamic Commemorative Structures, Notes and
Documents.” Ars Orientals 6 (1966): 7–46.
Green, Nile. Making Space: Sufis and Settlers in Early Modern India. New Delhi:
Green, Nile. “Introduction: The Frontiers of the Persianate World (c. 800–1900),”
in The Persianate World: The Frontiers of a Eurasian Lingua Franca (ed. Nile
Green), 1–71. Oakland, CA: University of California Press, 2019.
Grenet, Frantz. “The Nomadic Element in the Kushan Empire (1st–3rd Century
ad).” Journal of Central Eurasian Studies 3 (2012): 1–22.
Habib, Irfan. “Formation of the Sultanate Ruling Class of the Thirteenth
Century,” in Medieval India 1: Researches in the History of India 1200–1750 (ed.
Irfan Habib), 1–21. Delhi: Oxford University Press, 1992.

382 Iran to India
Habib, Irfan. “Review of M. A. Asif, A Book of Conquest: The Chachnama and

Muslim Origins in South Asia (Cambridge, MA: Harvard University Press),
2016.” Studies in People’s History 4(1) (2017): 105–117.
Haig, Malcolm Robert. The Indus Delta Country: A Memoir, Chiefly on Its Ancient
Geography and History, with Three Maps. London and Karachi: Kegan Paul,
Trench, Trubner and Indus Publications, 1894.
Hambly, Gavin. “Who Were the Chihilgānī, the Forty Slaves of Sul†ān Shams
Al-Dīn Iltutmish of Delhi?” Iran 10 (1972): 57–62.
Hanifi, Shah Mahmud. “Kāka®,” in Encyclopaedia of Islam, 2nd edn. (eds.
P. Bearman, Th. Bianquis, C. E. Bosworth, E. van Donzel, and W. P.
Haq, S. Moinul. “Shaykh Nizam Al-Din Awliya,” in Studies in Indian Culture:
Dr. Ghulam Yazdani Commemoration Volume (ed. A. K. Sherwani), 196–202.
Hyderabad: Maulana Abul Kalam Oriental Research Institute, 1966.
Hardy, Adam. Theory and Practice of Temple Architecture in Medieval India:
Bhoja’s Samaranganasutradhara and the Bhojpur Line Drawings. New Delhi:
Indira Gandhi National Centre for the Arts, 2015.
Hardy, Peter. Historians of Medieval India: Studies in Indo-Muslim Historical
Writing. London: Luzac, 1960.
Hardy, Peter. “Growth of Authority over a Conquered Political Elite: Early Delhi
Sultanate as a Possible Case Study,” in Kingship and Authority in South Asia,
216–241. Delhi: Oxford University Press, 1998.
Hartmann, Angelika. “Al-Suhrawardi, Shihab al Din Abu Hafs Umar,”
in Encyclopaedia of Islam, 2nd edn. (eds. P. Bearman, Th. Bianquis, C. E.
Bosworth, E. van Donzel, and W. P. Heinrichs). Leiden: Brill online, 2008.
Hegewald, Julia A. B. “The International Jaina Style? Maru–Gurjara Temples
under the Solankis, throughout India and in the Diaspora.” Ars Orientalis 45
(2015): 115–140.
Hillenbrand, Robert. Islamic Architecture. New York: Columbia University Press,
1994.
Hillenbrand, Robert. “The Seljuq Monuments of Turkmenistan,” in The Seljuqs:
Politics, Society and Culture (eds. Christian Lange and Songül Mecit), 277–308.
Hillenbrand, Robert. “Islamic Monumental Inscriptions Contextualized:
Location, Content, Legibility and Aesthetics.” Beiträge Zur Islamischen Kunst
und Archäologie 3 (2012): 13–38.
Hillenbrand, Robert. “Architecture and Politics: The North and South Dome
Chambers of the Isfahan Jami’,” in The Age of the Seljuqs (eds. Edmund
Herzig and Sarah Stewart), The Idea of Iran, 6:148–173. London: I. B. Tauris,
2015.
Hodgson, Marshall G. S. The Venture of Islam, Conscience and History in a World
Civilization, vol. 2: The Expansion of Islam in the Middle Periods. Chicago:
University of Chicago Press, [1974] 1977.
Holdich, Thomas, Sir (Colonel). The Gates of India, Being an Historical Narrative.
London: Macmillan, 1910.
Horn, Paul. “IX. – Muhammadan Inscriptions from the Suba of Dihli.” Epigraphia
Indica 3 (1894a): 130–159.

Horn, Paul. “Muhammadan Inscriptions from the Suba of Dihli No. II.”
Epigraphia Indica 3 (1894b): 424–437.
Horovitz, J. “The Inscriptions of Muhammad Ibn Sam, Qutbuddin Aibeg and
Iltutmish.” Epigraphia Indo-Moslemica 12 (1911): 12–34.
Hosain, Hidayet. “Hudjwīrī, Abu ’l-Óasan ‘Alī b. ‘Uthmān b. ‘Alī al-Ghaznawī
al-Djullābī,” in Encyclopaedia of Islam, 2nd edn. (eds. P. Bearman, Th.
Bianquis, C. E. Bosworth, E. van Donzel, and W. P. Heinrichs). Leiden: Brill
online, 2012.
Hua, T. “The Muslim Qarakhanids and Their Invented Ethnic Identity,” in
Islamisation de l’Asia Centrale (ed. Étienne de la Vassière), 339–350. Paris:
Association pour l’avancement des études iraniennes, 2008.
Husain, Muhammad Ashraf. A Record of All the Quranic and Non-Historical
Epigraphs on the Protected Monuments in the Delhi Province, Memoirs of the
Archaeological Survey of India 47. Calcutta: Government of India Central
Publication Branch, 1936.
Hüttel, Hans-Georg and Ulambayar Erdenebat. Karabalgasun and Karakorum
– Two Late Nomadic Urban Settlements in the Orkhon Valley. Archaeological
Excavation and Research of the German Archaeological Institute (DAI) and the
Mongolian Academy of Sciences 2000–2009, trans. Vincent and Eva Chandler.
Ulan Bator: Botschaft der Bundesrepublik Deutschland, 2010.
Inden, Ronald. “Hierarchies of Kings in Early Medieval India.” Contributions to
Indian Sociology 15(1/2) (1981): 99–125.
Inden, Ronald. “Introduction: From Philogical to Dialogical Texts,” in Querying
the Medieval: Texts and the History of Practices in South Asia, 3–28. Oxford:
Jackson, Peter. “Jala Al-Din, the Mongols, and the Khwarazmian Conquest of the
Panjab and Sind.” Iran 28 (1990): 45–54.
Jackson, Peter. “Jalāl Al-Dīn, the Mongols, and the Khwarazmian Conquest of
the Panjāb and Sind.” Iran 28 (1990): 45–54.
Jackson, Peter. The Delhi Sultanate: A Political and Military History. Cambridge:
Jackson, Peter. The Mongols and the Islamic World: From Conquest to Conversion.
New Haven, CT: Yale University Press, 2017.
Jackson, P. and P. A. Andrews. “Lāhawr,” in Encyclopaedia of Islam, 2nd edn.
Jain-Neubauer, Jutta. The Stepwells of Gujarat in Art-Historical Perspective. New
Delhi: Abhinav Publications, 1981.
Kalus, Ludwig. Thesaurus d’Épigraphie Islamique, online. Geneva: Fondation
Max van Berchem, 2017, available at: http://epigraphie-islamique.org/epi/
texte_acceuil.html.
Kaplony, A. “The Conversion of the Turks of Central Asia to Islam as Seen by
Arabic and Persian Geography: A Comparative Perspective,” in Islamisation de
l’Asia Centrale (ed. Étienne de la Vassière), 277–296. Paris: Association pour
l’avancement des études iraniennes, 2008.
Karamustafa, Ahmet T. God’s Unruly Friends: Dervish Groups in the Islamic Later
Middle Period, 1200–1550. Salt Lake City: University of Utah Press, 1994.

384 Iran to India
Karamustafa, Ahmet T. God’s Unruly Friends: Dervish Groups in the Islamic Middle
Period 1200–1550. Oxford: Oneworld, 2007.
Khazeni, Arash. Tribes and Empire on the Margins of Nineteenth-century Iran,
Publications on the Near East, University of Washington. Seattle: University
of Washington Press, 2009.
Khazeni, Arash. “Herat i. Geography,” in Encyclopaedia Iranica, online edition,
2012a.
Khazeni, Arash. “Through an Ocean of Sand: Pastoralism and the Equestrian
Culture of the Eurasian Steppe,” in Water on Sand: Environmental Histories
of the Middle East and North Africa (ed. Alan Mikhail), 133–158. New York:
Oxford University Press, 2012b.
Khismatulin, Alexey. “Two Mirrors for Princes Fabricated at the Seljuq Court:
Nizam al-Mulk’s Siyar al-Muluk and al-Ghazali’s Nasihat al-Muluk,” in The
Age of the Seljuqs, vol. 6: The Idea of Iran (eds. Edmund Herzig and Sarah
Stewart), 94–130. London: I. B. Tauris, 2015.
Kielhorn, F. “The Chahamanas of Naddula.” Epigraphia Indica IX (1907/8): 62–83.
Kimball, Fiske. “The Sasanian Building at Tepe Hissar,” in Excavations at Tepe
Hissar, Damghan, 347–350, figures 176, 177. Philadelphia, PA: University of
Pennzylvania Press for the University Museum, 1937.
King, David A. “Astronomical Alignments in Medieval Islamic Religious
Architecture,” eds. Bill M. Boland, Justine Cullinan, and Noemi E. Varas.
Annals of the New York Academy of Sciences 391 (1982): 303–312.
King, David A. Astronomy in the Service of Islam. Aldershot: Variorum, 1993.
King, David A. “The Orientation of Medieval Islamic Religious Architecture and
Cities.” Journal for the History of Astronomy 26 (1995): 253–274.
Knysh, Alexander. Sufism: A New History of Islamic Mysticism. Princeton, NJ:
Princeton University Press, 2017.
Kröger, Jens. Sasanidischer Stuckdekor, vol. 5: Baghdader Forschungen. Mainz am
Rhein: Verlag Philipp von Zabern, 1982.
Kröger, Jens. “Stucco Decoration,” in Encyclopaedia Iranica, online edition, July
2005.
Kumar, Sunil. “The Value of Ādāb Al-Mulūk as a Historical Source: An Insight
into the Ideals and Expectations of Islamic Society in the Middle Period (ad
945–1500).” Indian Economic and Social History Review 22(3) (1985): 307–327.
Kumar, Sunil. “Qutb and Modern Memory,” in The Partitions of Memory: The
Afterlife of the Division of India (ed. Suvir Kaul), 141–182. Bloomington:
Indiana University Press, 2002.
Kumar, Sunil. The Present in Delhi’s Pasts. Delhi: Permanent Black, 2002.
Kumar, Sunil. “Service, Status, and Military Slavery in the Delhi Sultanate:
Thirteenth and Fourteenth Centuries,” in Slavery and South Asian History
(eds. Indrani Chatterjee and Richard Maxwell Eaton), 83–114. Bloomington:
Indiana University Press, 2006.
Kumar, Sunil. The Emergence of the Delhi Sultanate, 1192–1286. New Delhi:
Laleh, Haeedeh, Abolfazl Mokarramifar, and Zahra Lorzadeh. “Le paysage
urbain de Nishapur,” in Greater Khorasan: History, Geography, Archaeology and
Material Culture (ed. Rocco Rante), 115–124. Berlin: De Gruyter, 2015.

Lambton, Ann K. S. State and Government in Medieval Islam, London Oriental

Series, vol. 36. London: RoutledgeCurzon, [1981] 2006.
Lambton, Ann K. S. and Janine Sourdel-Thomine. “Isfahan,” in Historic Cities of
the Islamic World (ed. C. Edmund Bosworth). Leiden: Brill, 2008.
Le Strange, Guy. The Lands of the Eastern Caliphate: Mesopotamia, Persia, and
Central Asia from the Moslem Conquest to the Time of Timur. Cambridge:
Levi, Scott C. The Indian Diaspora in Central Asia and Its Trade, 1550–1900, Brill’s
Inner Asian Library. Leiden: Brill, 2002.
Lewis, Franklin. “Reading, Writing and Recitation: Sanā’i and the Origins of the
Persian Ghazal,” PhD dissertation, University of Chicago, 1995.
Litvinsky, Boris. “Ajina Tepe,” in Encyclopaedia Iranica, online edition, [1984]
2011.
Litvinsky, Boris. “Buddhism iv. Buddhist Sites in Afghanistan and Central Asia,”
in Encyclopaedia Iranica, online edition, 2010.
Makdisi, George. “Madrasa and University in the Middle Ages.” Studia Islamica 32
(1970): 255–264.
Makdisi, George. “Law and Traditionalism in the Institutions of Learning
of Mediaeval Islam,” in Theology and Law in Islam (ed. Gustave E. von
Grunebaum), 75–88. Wiesbaden: O. Harrassowitz, 1971.
Makdisi, George. The Rise of Colleges: Institutions of Learning in Islam and the West.
Malagaris, George. “Warfare and Environment in Medieval Eurasia: Turkic
Frontiers at Dandanqan, Somnath and Manzikert,” in Turkish History and
Culture in India (eds. A. C. S. Peacock and Richard P. McClary), 25–55.
Mani, B. R. Delhi: Threshold of the Orient (Studies in Archaeological Excavations).
New Delhi: Aryan Books International, 1997.
Meisami, Julie Scott. “The Past in Service of the Present: Two Views of History
in Medieval Persia.” Poetics Today 14(2) (1993): 247–275.
Meisami, Julie Scott. Persian Historiography to the End of the Twelfth Century.
Meisami, Julie Scott. “Palaces and Paradises: Palace Description in Medieval
Persian Poetry,” in Islamic Art and Literature (eds. Oleg Grabar and Cynthia
Robinson), 21–54. Princeton, NJ: Markus Wiener, 2001.
McClary, Richard P. Medieval Monuments of Central Asia, Edinburgh Studies in
Islamic Art. Edinburgh: Edinburgh University Press, 2020.
Meister, Michael W. “Style and Idiom in the Art of Uparamala.” Muqarnas 10
(1993): 344–354
Meister, Michael W. “Temples along the Indus.” Expedition 38(3) (1996): 41–54.
Meister, Michael W. “Gandhara–Nagara Temples of the Salt Range and Indus.”
Kala, The Journal of the Indian Art History Congress 4 (1997/8): 45–52.
Meister, Michael W. “Temples of the Salt Range,” in Religion, Ritual, and Royalty,
132–139. Jaipur: Rawat Publication, 1999.
Meister, Michael W. “Pattan Munara : Minar or Mandir ?” in Hari Smriti Studies
on Art Archaeology and Indology (ed. Banerji Arundhati), 113–121. New Delhi-
110002: Kaveri Books, 2006.

386 Iran to India
Meister, Michael W. “Exploring Kafirkot: When Is a Rose Apple Not a Rose.”

Pakistan Heritage 1 (2009): 109–128.
Meister, Michael, Abdur Rehman, and Farid Khan. “Discovery of a New Temple
on the Indus.” Expedition 42(1) (2000): 37–46.
Melchert, Christopher. “Sufis and Competing Movements in Nishapur.” Iran 39
(2001): 237–247.
Melchert, Christopher. “Al-Bukhārī,” in Encyclopaedia of Islam, 3rd end. (eds.
Kate Fleet, Gudrun Krämer, Denis Matringe, John Nawas, and Everett
Rowson), Leiden: Brill online, 2012.
Melikian-Chirvani, Assadullah Souren. “L’évocation littéraire du boudhisme dans
l’Iran musulman,” in Le monde iranien et l’Islam, sociétés et cultures (ed. Jean
Aubin), II:59–92. Geneva: Librarie Droz, 1974.
Michailidis, Melanie. “In the Footsteps of the Sasanians: Funerary Architecture
and Bavandid Legitimacy,” in Persian Kingship and Architecture: Strategies of
Power in Iran from the Achaemenids to the Pahlavis (eds. Sussan Babaie and
Grigor Talinn), 135–174. London: I. B. Tauris, 2015.
Miles, George C. “Epigraphical Notice.” Ars Islamica 6(1) (1939): 11–15.
Nazim, M. “The Pand-Namah of Subuktigan.” Journal of the Royal Asiatic Society
of Great Britain & Ireland (1933): 605–628.
Neelis, Jason. Early Buddhist Transmission and Trade Networks: Mobility and
Exchange within and beyond the Northwestern Borderlands of South Asia, vol. 2:
Dynamics in the History of Religion. Leiden: Brill, 2011.
Nizami, K. A. Some Aspects of Religion and Politics in India during the Thirteenth
Century. Delhi: Mohammad Ahmad Idarah-i Adabiyat-i Delli, [1961] 1978
(reprint).
Nizami, K. A. On History and Historians of Medieval India. New Delhi: Munshiram
Manoharlal, 1983.
Nizami, K. A. “Popular Movements, Religious Trends and Sufi Influence on the
Masses in the Post-‘Abbasid Period,” in The Age of Achievement: AD 750 to the
End of the Fifteenth Century – Part One: The Historical, Social and Economic
Setting, vol. 4: History of Civilizations of Central Asia (eds. M. S. Asimov and
C. E. Bosworth), 365–379. Paris: UNESCO Publishing, 1998.
Nizami, K. A. “Čishtī, Khwādja Mu‘īn al-Dīn Óasan,” in Encyclopaedia of Islam,
2nd edn. (eds. P. Bearman, Th. Bianquis, C. E. Bosworth, E. van Donzel, and
W. P. Heinrichs). Leiden: Brill online, 2011a.
Nizami, K. A. “Čishtiyya,” in Encyclopaedia of Islam, 2nd edn. (eds. P. Bearman,
Th. Bianquis, C. E. Bosworth, E. van Donzel, and W. P. Heinrichs). Leiden:
Brill online, 2011b.
O’Kane, Bernard. The Appearance of Persian on Islamic Art, vol. 4, Biennial
Ehsan Yarshater Lecture Series. New York: Persian Heritage Foundation,
2009.
Patel, Alka. Building Communities in Gujarāt: Architecture and Society during
the Twelfth through Fourteenth Centuries, Brill’s Indological Library, vol. 22.
Leiden: Brill, [2004] 2007.
Patel, Alka. “A Note on Mahmud Ghaznavi, Somanatha, and the Building of a
Reputation,” in South Asian Archaeology 2001. (ed. Vincent Lefèvre), 599–606.
Paris: Editions Rechèrche sur les Civilisations, 2005.

Patel, Alka. “Revisiting the Term ‘Sultanate’,” in The Architecture of the Indian
Sultanates (eds. Abha Narain Lambah and Alka Patel), 8–12. Mumbai: Marg
Publications, 2006.
Patel, Alka. “Hind Wa Sind: Textual and Material Evidence of Muslim
Communities in Seventh- and Eighth-Century South Asia,” paper presented
at the Association for Asian Studies, Philadelphia, PA, 2010.
Patel, Alka. “Architectural Cultures and Empire: The Ghurids in Northern India
(c. 1192–1210).” Bulletin of the Asia Institute 21 ([2007] 2011): 35–60.
Patel, Alka. “The Rehmā~a-Prāsāda Abroad: Masjid-i Sangī of Larwand (Farah
Province, Afghanistan),” in Prāsāda-Niddhī: Temple Architecture and Sculpture
of South and Southeast Asia, Essays in Honour of Professor M. A. Dhaky (eds.
P. P. Dhar, Gerd Mevissen, and Devangana Desai), 84–99. New Delhi: Aryan
Books, 2015.
Patel, Alka. “Text as Nationalist Object: Modern Persian-Language Historiography
on the Ghurids (c. 1150–1215),” in India and Iran in the Longue Durée (eds. Alka
Patel and Touraj Daryaee), 143–166. Irvine, CA: Jordan Center for Persian
Studies, 2017.
Patel, Alka. “Periphery as Center: The Ghurids between the Persianate and Indic
Worlds,” in The Idea of Iran: From Saljuq Collapse to Mongol Conquest (eds.
Sarah Stewart and Edmund Herzig), 29–53. London: I. B. Tauris, 2018.
Paul, J. “Islamizing Sufis in Pre-Mongol Central Asia,” in Islamisation de l’Asia
Centrale (ed. Étienne de la Vassière), 297–318. Paris: Association pour l’avance-
ment des études iraniennes, 2008.
Peacock, A. C. S. Early Seljuq History: A New Interpretation, Routledge Studies in
the History of Iran and Turkey. New York: Routledge, 2010.
Peacock, A. C. S. The Great Seljuk Empire, Edinburgh History of the Islamic
Empires. Edinburgh: Edinburgh University Press, 2015.
Peacock, A. C. S. “The Great Age of the Seljuks,” in Court and Cosmos: The Great
Age of the Seljuqs (eds. Sheila R. Canby, Deniz Beyazit, and Martina Rugiadi),
2–35. New York and New Haven: Metropolitan Museum of Art and Yale
Peacock, A. C. S. “Firdawsi’s Shahnama in Its Ghaznavid Context.” Iran 56(1)
(2018): 2–12.
Pedersen, J., G. Makdisi, Munibur Rahman, and R. Hillenbrand. “Madrasa,” in
Encyclopaedia of Islam, 2nd edn. (eds. Th. Bianquis, C. E. Bosworth, E. van
Donzel, W. P. Heinrichs, and P. Bearman). Leiden: Brill online, 2012.
Pons, Jessie. “Kushan Dynasty ix. Art of the Kushans,” in Encyclopaedia Iranica,
online edition, 2016.
Prasad, Pushpa. Sanskrit Inscriptions of Delhi Sultanate, 1191–1526. Delhi: Oxford
Rabbat, Nasser O. The Citadel of Cairo: A New Interpretation of Royal Mamluk
Architecture. Islamic History and Civilization, vol. 14. Leiden: Brill, 1995.
Raikes, Captain Stanley Napier. Memoir on the Thurr and Parkur Districts of Sind,
1856, vol. LIV: Selections from the Records of the Bombay Government, New
Series. Bombay: Education Society’s Press, Byculla, 1859.
Rante, Rocco. “‘Khorasan Proper’ and ‘Greater Khorasan’ within a Politico-
Cultural Framework,” in Greater Khorasan: History, Geography, Archaeology

388 Iran to India
and Material Culture (ed. Rocco Rante), 9–26. Berlin: De Gruyter,

2015.
Rante, Rocco and Annabelle Collinet. Nishapur Revisited: Stratigraphy and
Ceramics of the Qohandez. Oxford: Oxbow Books, 2013.
Ray, H. C. The Dynastic History of Northern India. 2nd edn. 2 vols. Delhi:
Munshiram Manoharlal Publishers, 1973.
Ray, Himanshu Prabha. “Maritime Archaeology of the Indian Ocean: An
Overview,” in Tradition and Archaeology: Early Maritime Contacts in the
Indian Ocean (eds. Himanshu Prabha Ray and Jean-François Salles), 1998th
edn., 1–10. Delhi: Manohar, 1996.
Ray, Himanshu Prabha. “The Beginnings: The Artisan and the Merchant in
Early Gujarat, Sixth–Eleventh Centuries.” Ars Orientalis 34 ([2004] 2007):
39–61.
Répertoire chronologique d’épigraphie arabe. Cairo, Egypt: Institut français
d’archéologie orientale du Caire: vol. V (Tome cinquième), années 354 à 386 de
l’Hégire, 1934 ; vol. VI (Tome sixième), années 386 à 425 de l’Hégire, 1935 ; vol.
IX (Tome neuvième), années 550 à 601 de l’Hégire, 1937.
Reu, Sāhityāchārya Pt. Bisheshwar Nath. “The Gāha∂avālas of Kanauj from about
V.S. 1125 (A.D. 1068) to about V.S. 1280 (A.D. 1223).” Journal of the Royal
Asiatic Society of Great Britain and Ireland 1 (1932): 1–21.
Rezakhani, Khodadad. ReOrienting the Sasanians: East Iran in Late Antiquity.
Rezakhani, Khodadad. “Pangul and Banji, Zhulad and Fulad: A Note on the
Genealogy of the Shanasbid Amirs of Ghur,” in Dinars and Dirhams: Festschrift
in Honor of Michael L. Bates (eds. Touraj Daryaee, Judith Lerner, and Virginie
Rey). Irvine, CA: Jordan Center for Persian Studies, 2020.
Rosenfield, John M. Dynastic Arts of the Kushans, 3rd edn. Berkeley: University of
California Press, [1967] 1993.
Ross, E. Denison. “The Genealogies of Fakhruddin Mubarakshah,” in A Volume
of Oriental Studies Presented to Edward G. Browne on His 60th Birthday (7
February 1922) (eds. Thomas Walker Arnold and Reynold Alleyne Nicholson),
392–413. Cambridge: Cambridge University Press, 1922.
Ruthven, Malise and Azim Nanji. Historical Atlas of Islam. Cambridge, MA:
Harvard University Press, 2004.
Sarkar, Jagadish Narayan. History of History-Writing in Medieval India. Calcutta:
Ratna Prakashan, 1977.
Schmidt, Erich F. Excavations at Tepe Hissar, Damghan. Philadelphia, PA:
University of Pennsylvania Press for the University Museum, 1937.
Schwarz, Florian. Sylloge Numorum Arabicorum Tübingen. Balh und die
Landschaften am oberen Oxus, XIV c Hurasan III. Tübingen: Ernst Wasmuth
Verlag, 2002.
Sen, Sudipta. “Historian as Witness: Ghulam Husain Tabatabai and the Dawning
of British Rule in India,” in India and Iran in the Longue Durée (eds. Alka Patel
and Touraj Daryaee), 103–124. Irvine, CA: Jordan Center for Persian Studies,
2017.
Sevim, A. and C. E. Bosworth. “The Seljuqs and the Khwarazm Shahs,” in History
of Civilizations of Central Asia, vol. IV: The Age of Achievement: AD 750 to the

End of the Fifteenth Century – Part Two: The Achievements (eds. C. E. Bosworth
and M. S. Asimov), 1st Indian edn., 145–176. New Delhi: Motilal Banarsidass
Publishers, 1999.
Sharma, Sunil. Persian Poetry at the Indian Frontier: Masʻûd Saʻd Salmân of
Lahore. Permanent Black Monographs, Opus 1 Series. New Delhi: Permanent
Black, 2000.
Shokoohy, Mehrdad and Natalie H. Shokoohy. “The Architecture of Baha Al-Din
Tughrul in the Region of Bayana, Rajasthan.” Muqarnas 4 (1987): 114–132.
Shokoohy, Mehrdad and Natalie H. Shokoohy. Bhadreśvar: The Oldest Islamic
Monuments in India, Studies in Islamic Art and Architecture, 2. Leiden: Brill,
1988.
Siddiqui, Iqtidar Husain. Authority and Kingship under the Sultans of Delhi
(Thirteenth–Fourteenth Centuries). Delhi: Manohar, 2006.
Siddiqui, Iqtidar Husain. Indo-Persian Historiography up to the Thirteenth Century.
Delhi: Primus Books, 2010.
Smith, Frederick. “Trees and Plants,” in Brill’s Encyclopedia of Hinduism (eds.
Knut A. Jacobsen, Helene Basu, Angelika Malinar, and Vasudha Narayanan).
Smith, Myron Bement. “The Manars of Isfahan.” Athar-e Iran I(2) (1936): 313–358.
Smith, Myron Bement. “Epigraphical Notice.” Ars Islamica 6(1) (1939): 11–15.
Sobieroj, F. “Suhrawardiyya,” in Encyclopaedia of Islam, 2nd edn. (eds. P. Bearman,
Th. Bianquis, C. E. Bosworth, and W. P. Heinrichs). Leiden: Brill online, 2012.
Sourdel-Thomine, Janine. “Les décors de stuc dans l’est iranien à l’époque sal-
gūqide.” Akten des vierundwanzigsten international Orientalisten-Kongresses 44
(1957): 342–344.
Sourdel-Thomine, Janine. “Réflexions sur la diffusion de la madrasa en orient du
XIe au XIIIe siècle.” Révue des études islamiques 44 (1976): 165–184.
Spooner, Brian. “Baluchistan i. Geography, History and Ethnography,” in
Encyclopaedia Iranica, online edition, [1988] 2010.
Spooner, Brian. “Epilogue: The Persianate Millennium,” in The Persianate World:
The Frontiers of a Eurasian Lingua Franca (ed. Nile Green), 301–316. Oakland,
CA: University of California Press, 2019.
Subtelny, Maria Eva. “The Symbiosis of Turk and Tajik,” in Central Asia in
Historical Perspective, 45–61. Cambridge, MA and Boulder, CO: Harvard
University Russian Research Center and Westview Press, 1994.
Tetley, G. E. The Ghaznavid and Seljuq Turks: Poetry as a Source for Iranian
History. London: Routledge Taylor & Francis, 2009.
Thapar, Romila. Somanatha: The Many Voices of a History. New Delhi: Penguin/
Viking, 2004.
Tor, D. G. “‘Sovereign and Pious’: The Religious Life of the Great Seljuq Sultans,”
in The Seljuqs: Politics, Society and Culture (eds. Christian Lange and Songül
Mecit), 39–62. Edinburgh: Edinburgh University Press, 2011.
Trimingham, J. Spencer. The Sufi Orders in Islam. Oxford: Clarendon Press, 1971.
Tritton, A. S. Materials on Muslim Education in the Middle Ages. London: Luzac,
1957.
Trivedi, Harihar Vitthal. Chaulukyan Inscriptions. Madhya Pradesh: Commissioner
of Archaeology, Archives and Museums, 1994.

390 Iran to India
Tye, Robert and Monica Tye. Jitals: A Catalogue and Account of the Coin
Denomination of Daily Use in Medieval Afghanistan and North West India. Isle
of South Uist: Robert Tye, 1995.
Van Renterghem, Vanessa. “Baghdad: A View from the Edge on the Seljuq
Empire,” in The Age of the Seljuqs, vol. 6: The Idea of Iran (eds. Edmund Herzig
and Sarah Stewart), 74–93. London: I. B. Tauris, 2015.
Vàsàry, Istvàn. “Two Patterns of Acculturation to Islam: The Qarakhanids versus
the Ghaznavids and Seljuqs,” in The Age of the Seljuqs, vol. 6: The Idea of Iran
(eds. Edmund Herzig and Sarah Stewart), 9–28. London: I. B. Tauris, 2015.
Welch, Anthony and H. Crane. “The Tughluqs: Master Builders of the Delhi
Sultanate.” Muqarnas I (1983): 123–166.
Wilkinson, Tony J. “Introduction,” in Landscapes of the Islamic World: Archaeology,
History, and Ethnography (eds Stephen McPhillips and Paul D. Wordsworth),
1–16. Philadelphia: University of Pennsylvania Press, 2016.
Wordsworth, Paul D. “Merv on Khorasanian Trade Routes from the 10th–13th
Centuries,” in Greater Khorasan: History, Geography, Archaeology and Material
Culture (ed. Rocco Rante), 51–62. Berlin: De Gruyter, 2015.
Yusofi, Gholam-Hosayn. “Chahar Maqala,” in Encyclopaedia Iranica, online
edition, 1990.
Zysow, Aron. “Two Unrecognized Karrami Texts.” Journal of the American
Oriental Society 108(4) (1988): 577–587.
Zysow, Aron. “Karamiyya,” in Encyclopaedia Iranica, online edition, [2011] 2012.

Index
Note: bold indicates illustrations
‘Abbas ibn Shish, 90, 195, 238 ribat of ‘Ali ibn Karmakh, Multan,
Abi al-Fath ibn Muhammad, 48–9 304–5, 306–8
Abu Ja‘far Muhammad ibn ‘Ali, 52 Robat Sharaf, 57
Abu Nasir Ahmad ibn Fadl, 48 Sarkhushak, 160
Aditya temple, Multan, 282, 300 tomb of Ahmad Kabir, Multan, 304,
administrative structures, 48, 83, 116, 149, 316
150 tomb of Sadan Shahid, Multan, 304, 313
Afshin, 194 trefoil arches, 287, 288–90, 303–5,
agriculture, 18, 41, 61, 84, 87, 94–5, 154, 306–8, 313, 316
292–3 architectural treatises, 298, 303, 306
Ahangaran, 41, 81, 84–7, 86, 95, 95 Arhai Din-ka Jhonpra mosque, Ajmer, 9
Ajmer, 9, 135–6, 143, 146 art-architectural historiography, 17–18,
‘Ala’ al-Din Husain, 62–5, 132–5, 154, 171, 81–2, 148, 241
173, 194–5, 208, 238, 241–2, 255–7, artisans, 23, 27, 94–6, 108–9, 143–4, 191,
262–3, 300–1, 322 206, 221, 281, 296, 298, 302–3, 306,
‘Ala’ al-Din Muhammad, 65 327–8, 339
Alana Valley, 92 Asif, Manan Ahmed, 42, 79
Ali, Taj, 278 Astarabad, 65
‘Ali ibn Abi Talib, 43 Audience Hall, Lashkari-Bazar, 259, 263,
Allchin, F. Raymond, 150 265–7, 266–8
Allen, Terry, 253 Awbeh, 86
Alptigin, 16, 221 Azeem, Rizwan, 323
Ambh Sharif temple, Panjab, 290–1 Azhd Zahhak, 42–3
Amir Banji ibn Naharan, 43, 107, 108, 116
Anahilavada-pattana, 67 Babur, 8
Anandapala, 285 Bactria, 4–5, 148, 304
Anatolia, 8, 59, 116, 292 Badghis, 82, 86
Anooshahr, Ali, 42, 279 Baghdad, 59, 107, 116
anthropomorphism, 198, 219 Baha’ al-Din, 62, 64, 134, 135, 154–5, 194, 238
arabesque ornament, 50, 157, 163, 244, 247, Baha’ al-Din Tughril, 65
259–60, 304, 323 Bahram Shah, 62–3, 64, 109, 194, 208, 242
arch motifs, 98, 103, 104–5, 212–13, 216 Bahram Shah minar, Ghazna, 58, 109–11,
arches 110; see also minarets of Ghazni
Bust monumental arch, 7, 249, 251–4, Baihaqi, 41
252–3, 264 baked brick, 25, 26, 50, 91, 100, 109, 139,
Chisht domed structures, 212, 212–13, 142–3, 250, 299, 305, 318
215 Baker, P. H. B., 150
Gandhara–Nagara school, 287, 288–90 Bakhtyari tribe, 240

392 Iran to India
Balasaghun region, 45–6 Bust monumental arch, 7, 249, 251–4,

Balkh, 26, 52, 54, 61, 82, 109–11, 115, 133, 252–3, 264
148, 153, 169, 190
Ball, Warwick, 57, 88, 90, 92 Casimir, Michael, 200, 203, 240
Baluchistan, 278, 283, 284, 289–98, 293, cemeteries, 171–2, 172, 320; see also burial
302, 323–4, 327 mounds; graves; tombs
Bamiyan, 3, 5–6, 54, 82–3, 86, 132–3, 139, ceramics, 139, 155, 158, 162–3, 172, 242
147, 147–69, 149, 151–3, 156–68, ceremony, 54, 61, 106, 192, 206–7, 253,
173–5, 191–2, 238–40, 339 265–7
Bamiyan Shansabanis, 19, 26, 61–3, 82, Chach-nama (Kufi), 79
132, 154–5, 165–9, 171, 175, 190–2, 238, Chaghcharan, 84, 86, 134
267–8, 279, 281, 338, 339 Chagri, 48
Banbhore see Dibul-Banbhore Chahar Maqala (Nizami ‘Aruzi), 154
Bandian, 100 Chaulukyas, 67, 296
banditry, 41 Chehel Abdal, 84
Bara Gumbad mosque, Delhi, 310 Chehel Burj, 161
barbarians, 18, 19–20, 21, 108 Chhoti masjid, Bhadreshvar, 295, 318
Barfak II, 156–9, 162–5, 167 Chihil Dukhtaran, Isfahan, 48–9, 50–2
Barfield, Thomas, 341 China, 10, 46
Barsiyan mosque, 260 Chinggis Khan, 155
Basseri tribe, 240 Chinggisids, 155
bastions, 67, 90, 151–2, 157, 159, 160 Chira, 320
Bhadreshvar, 295, 295–6, 319 Chisht, 81, 86, 133, 190–7, 210–17, 211–13,
Bhimadeva II, 67 215–16, 220, 221, 261, 309, 312
Bhodesar temple, Nagarparkar, 297 Chishtiyya, 215–17
Bistam, 65 Christianity, 45–6
Blair, Sheila, 199, 214, 221, 254, 310 cisterns, 139, 142
Bosworth, C. E., 41, 90, 114, 198 citadels, 28, 29, 156, 170, 192, 248–54,
Brahmanical Hinduism, 56–7, 282, 285, 249; see also fortifications; hilltop
289, 304 complexes
Bronson, Bennet, 19 city walls, 7, 248–9; see also fortifications
Buddha sculptures, 147, 148 coinage, 44, 61, 62, 63–4, 65, 80, 263, 264,
Buddhism, 1, 45–6, 56–9, 83, 109, 147–50, 340
157, 159, 167, 169–71, 239, 284–5, 287, commemoration, 67, 143, 146, 209–10,
304 218–19, 253–4, 282, 300–1, 316, 321,
Buddhist monasteries, 56–9, 59, 83, 322
148–50, 159, 169–71, 170, 239, 284, commerce, 3, 42, 47, 51, 62–3, 107, 113–16,
284–5 133–9, 142, 148–50, 153–4, 169–75,
building materials, 4, 23, 50, 90–1, 278, 222, 239–40, 279, 282–5, 289–90, 294,
287, 295, 299, 302, 305 299–300, 329; see also trade
al-Bukhari, 207 commercial architecture, 136, 139, 248, 294
Bulliet, Richard, 217 communication routes, 51, 175, 195, 197,
Burbank, Jane, 10, 19 220, 240, 279, 285, 323–4, 325
burial mounds, 3, 4, 133, 171–2, 172, 173; confederation, 5, 16, 40, 61–2, 116, 193,
see also graves; tombs 240, 241, 338, 340
burial practices, 171–2, 301 Congregational Mosques
Bust galleried well, 249–51, 250–1 Herat, 4, 11, 12, 211
Bust-Lashkari Bazar, 3, 7, 23–5, 54, 63, 90, Isfahan, 48, 49–50
133–4, 154, 164–5, 171–3, 192–3, 237–8, Jam Jaskars Goth, 294, 294–5
241–2, 247–67, 248–61, 265–70, 281, Multan, 282, 283
301, 308–9, 321 Cooper, Frederick, 10, 19

Index 393
corbeling, 161, 295, 296 Gandhara–Nagara school, 287–9,

corridors, 66, 153, 161, 287–9, 304–5 288–91, 305–6, 309, 327
cosmopolitanism, 23, 50, 59, 100, 103, 106, geometric ornament, 25, 50, 51, 99–100,
108–13, 143–6, 175, 191, 221–2, 289–92, 111, 144–5, 145, 151–2, 156–7, 157, 162,
323, 328 165, 247, 253, 255, 259, 292, 304, 315,
courts, 16, 21, 40, 42, 44–5, 60, 107, 146–7, 323
154–5, 192, 221, 222, 267–8, 341 Ghur architectural remains, 90, 91,
courtyards 96–9, 96–103, 102, 104, 109
Barfak II, 157, 158–9, 163 glazed tiles, 49, 50, 54, 247, 278, 302,
courtyard-ivan configuration, 51, 54, 157, 320–4, 322
158–9, 163, 164 Jam-Firuzkuh minar, 144–6, 145, 322
Mas‘ud III palace, Ghazna, 6, 243–4, Lal Mara Sharif tombs, 278, 302, 313,
262–3 320–4, 322, 327–8
Robat Sharaf, 51, 56 Mas‘ud III palace, Ghazna, 242–7,
Shahr-i Ghulghula, 163, 164 245–7, 323
South Palace, Lashkari Bazar, 255, 257, medallions, 25, 212–13
258, 263 mural paintings, 152, 156, 161, 267,
Cousens, Henry, 295 268
Crossley, Patricia, 18 overflowing pot motifs (purnaghata),
cruciform plans, 92, 94, 164 295, 305, 308, 310, 315–18, 318
cursive script, 194, 203, 209, 209–11, ribat of ‘Ali ibn Karmakh, Multan,
211–13, 213–14 302–3, 304–6, 306–8, 310–11
Robat Sharaf, 51, 57
Da Qadi minar, 99–100, 100, 108–12, 113, Saljuq architecture, 50, 98, 109
143, 191, 281, 339 Sasanian architecture, 100–3, 103–5
dadoes, 6, 103, 244, 246–7, 247, 257, 259, Shah-i Mashhad madrasa, 199, 202–3
263 South Palace, Lashkari Bazar, 255,
Damascus, 48 259–60, 260, 263, 267, 267–8
Dandanqan, battle of, 48, 61 stucco decoration, 100–3, 102–4, 111,
Danestama see Barfak II 157, 165, 212, 216, 242–3, 245–6, 251,
Darmashan, 84, 86 255, 259–60, 260, 267
Daulatabad minar, 52, 99, 102, 110–11 Sukkhur tombs, 302, 313–18, 318, 323,
Daykundi 82 327
de Bruijn, Thomas, 8 terracotta decoration, 25, 242–7, 292–3,
decoration 293, 324
arabesque ornament, 50, 157, 163, 244, Thamban Wari masjid, Jam Jaskars
247, 259–60, 304, 323 Goth, 295–6, 298
arch motifs, 98, 103, 104–5, 212–13, 216 tomb of Ahmad Kabir, Multan, 277–8,
Baluchistan tombs, 292–3, 293, 324, 327 302–4, 314, 316
Bamiyan hilltop complexes, 156–7, 157, tomb of Muhammad ibn Harun,
160, 161, 165 Lasbela, 324
Bust monumental arch, 251–4, 252–3 tomb of Salar Khalil, Juzjan, 24–6, 25
ceramics, 162–3 tomb of Sadan Shahid, Multan, 302–4,
Chisht domed structures, 212–13, 216 313
Da Qadi minar, 99–100, 100, 109–11 tomb of Shahzada Shaikh Husain, Bust,
epigraphic motifs, 247, 313, 315–16, 318, 24–5, 27
327 vegetal ornament, 144–5, 157, 162, 295,
figural decoration, 292–3 305
floral ornament, 50, 51, 98, 98–9, 101, zoomorphic motifs, 247, 304, 323
102, 104, 111, 157, 244, 247, 259–60, see also epigraphy
292, 304, 323 defensive structures see fortifications

394 Iran to India
Délégation Archéologique Française en funerary inscriptions, 204, 213–14

Afghanistan (DAFA), 161–2 Gandhara–Nagara school, 305–6, 309,
Delhi, 44, 109, 282, 310–12 327
Delhi Sultanate, 15 Hadith inscriptions, 193, 204, 206, 207,
Dhaky, M. A., 296 222
Dibul-Banbhore, 67, 289–90, 292, 313 historical inscriptions, 51, 144–5, 204,
domes 304, 307
Chisht domed structures, 190–7, 210–17, Jam-Firuzkuh minar, 143, 144–5, 190,
211–13, 215–16, 220, 221, 261 193, 197, 217–20
Congregational Mosque, Isfahan, 48, Kufic script, 25, 26, 49, 102, 111, 194,
49 202–3, 203, 205, 209–11, 214, 215,
corbeled domes, 295, 296 219, 244, 257, 304, 318, 319
domed cube tombs, 23–4, 24, 213, 278, legibility, 145, 204–5, 219
293, 302, 304, 314, 315, 317, 320, 320, Mas‘ud III minar, Ghazna, 208–9,
324, 326 208–10, 260
Gandhara–Nagara temples, 287 Mas‘ud III palace, Ghazna, 243–7,
hilltop complexes, 152, 160 257–9, 279
Saljuq architecture, 50 naskh script, 102, 111, 201, 202, 204,
Sarkhushak, 160 206, 211, 244, 304
Umayyad Mosque, Damascus, 48 and patronage, 192, 204–8, 210, 219–20,
doxologies, 26, 209, 213, 220, 222, 243, 257, 262, 264, 307–13
260, 261, 304, 307, 327 Qasimabad minar, 53
Dupree, Louis, 83 Qur’anic inscriptions, 26, 144–5, 193,
dwelling towers, 89, 90–4, 91, 92 197, 203–4, 207–14, 212–13, 215,
218–19, 222, 251–4, 252–3, 259–62,
Edwards, Holly, 23, 278, 315, 320 260, 302–13, 313–14, 327
enclosures see fortified enclosures; textile Qutb mosque, Delhi, 311
enclosures riba† of ‘Ali ibn Karmakh, Multan, 277,
endowments (waqf ), 263, 281, 282, 283, 302–3, 304–6, 306–8, 309
327, 340 Robat Sharaf, 51
epigraphy Saljuq architecture, 50, 254
Bara Gumbad mosque, Delhi, 310 Shah-i Mashhad madrasa, 193, 194–5,
Barsiyan mosque, 260 199, 201–10, 203–6, 253, 260, 307–8,
Bust monumental arch, 251–4, 252–3, 313, 339
264 Sharada script, 304
Chihil Dukhtaran, Isfahan, 51, 52 Shrine of Ibrahim, Bhadreshvar, 319
Chisht domed structures, 193, 194, Sin minar, 49
210–14, 211–13, 215–16, 261, 309, South Palace, Lashkari Bazar, 242,
312 254–62, 257, 263, 267, 308–9
Congregational Mosque, Herat, 211 thuluth script, 202, 203, 204
cursive script, 194, 203, 209, 209–11, tomb of Ahmad Kabir, Multan, 277,
211–13, 213–14 302–4, 309–10, 314, 316
and dating, 24, 26, 51, 109, 190, 194, 211, tomb of Iltutmish, Delhi, 311–12
254–5, 262, 277 tomb of Mir Sayyid Bahram,
Daulatabad minar, 52, 102 Kirminiyya, 310
doxologies, 26, 209, 213, 220, 222, 243, tomb of Sadan Shahid, Multan, 261,
257, 260, 261, 304, 327 302–4, 309, 312, 313
floriated Kufic script, 244 tomb of Salar Khalil, Juzjan, 24,
foliated Kufic script, 211 25–6
foundation inscriptions, 194–5, 203, tomb of Shahzada Shaikh Husain, Bust,
203, 210, 277 24

Index 395
Fakhr al-Din Mas‘ud, 62, 63, 132, 134, 154, Gharjistan, 84, 133, 190–210, 196, 200,
238, 268 202–6, 220, 238
Fakhr-i Mudabbir, 282 al-Ghazali, 154
Farah, 82, 87 Ghazna, 3, 6, 28, 53–68, 58, 109–11, 110,
Farah Rud, 87 132–4, 146, 154, 164–5, 171, 173–4,
Faryab, 82, 86 192–3, 208–9, 208–10, 219, 237–49,
Fatimids, 282, 300 243–7, 254–5, 260, 262–7, 279–82,
figural decoration, 292–3 301, 321–2, 338, 340
fire damage, 241–2 Ghazna Shansabanis, 19, 43, 61, 65, 174–5,
Firozkohis, 94, 95 190–1, 217, 237–41, 262–70, 277–98,
First Anglo-Afghan War, 29 301, 303, 313, 319–23, 325–8, 338, 340–1
Firuzkuh see Jam-Firuzkuh Ghaznavids, 5, 8, 15–17, 26, 41, 44, 46–8,
Firuzkuh Shansabanis, 19, 26, 42, 61–8, 84, 53–68, 84, 98, 109, 113–16, 132, 144,
132–47, 154–5, 167, 171–5, 190–7, 204, 154, 163–5, 171, 173, 192–3, 195–8,
210, 217–22, 238–41, 261–4, 279, 281, 206–10, 218, 221, 237, 240–2, 248–54,
313, 321, 338–9; see also Jam-Firuzkuh 257, 262–9, 278–82, 285, 299–301,
Flood, F. B., 17–18, 20, 197, 219 307–8, 316, 321–2, 325–7, 339–41
flood damage, 137, 139 Ghiyath al-Din, 4, 43, 62–7, 132, 135–7,
floral ornament, 50, 51, 98, 98–9, 101, 102, 155, 173–4, 194–8, 210–11, 214, 238,
104, 111, 157, 244, 247, 259–60, 292, 240, 254, 263, 322
304, 323 ghulams, 16, 53–4, 65, 79, 153–4, 221, 240,
floriated Kufic script, 244 311
foliated Kufic script, 211 Ghur, 1, 19–20, 27, 41–3, 46–7, 61, 80–117,
folk astronomy, 201, 214 81, 88–99, 102, 104, 132, 135, 139,
fortifications, 5, 41, 67, 81, 84, 87–98, 89, 150–4, 173, 174, 194, 198, 220–1, 238,
91–4, 96, 132–3, 136, 139, 140–1, 142, 240, 248, 281, 283, 338
150–67, 151–3, 156–68, 191–5, 248–9, Glatzer, Bernt, 200, 203, 240
294 glazed brick, 139
fortified enclosures, 80, 90–2 glazed tiles, 49, 50, 54, 247, 278, 302,
foundation inscriptions, 194–5, 203, 203, 320–4, 322
210, 277 Godard, André, 51
four-ivan plans, 164 Gol-Zarriun, battle of, 149
Fuladi, 148 graves, 3, 4, 133, 171–2, 172, 173, 321; see also
Funduqistan, 148 burial mounds; tombs
funerary inscriptions, 204, 213–14 Greco-Roman architecture, 287
grottoes, 150, 157, 160
Gandhara, 5, 283, 284, 284–9, 286–91, Gujarat, 67, 280, 289, 295, 296–8, 302,
304, 307 303, 316
Gandhara–Nagara school, 285–9,
288–91, 302, 303, 305–6, 309, 316, Hadiqat al-Haqiqa (Sana’i), 1
319, 327 Hadith inscriptions, 193, 204, 206, 207,
Ganga-Yamuna duab, 5, 67 222
Gar minar, 49 Halam ibn Shaiban, 282
garm-sir encampments, 87, 105, 106, 139, Hanafi madhhab, 198, 199, 201, 214, 217,
169, 194, 222, 237 219, 278–9
Gellner, Ernest, 47 Hari Rud, 41, 84–7, 85, 86, 94, 115, 132,
genealogies, xviii, 42–3 139, 167, 195–7
geometric ornament, 25, 50, 51, 99–100, Harun al-Rashid, 107
111, 144–5, 145, 151–2, 156–7, 157, 162, Helmand Province, 26, 82, 248
165, 247, 253, 255, 259, 292, 304, 315, Helmand River, 248, 267, 269–70
323 Hephthalites, 31, 81, 106, 149, 150, 153, 167

396 Iran to India
Herat, 4, 11–14, 30, 31, 52, 61, 63, 65, 81–2, ivans, 12, 51, 54, 56, 157, 158–9, 163, 164,
86, 115, 133, 169, 173, 175, 190, 195, 259
214–15, 220 ‘Izz al-Din, 61, 62, 132, 134
Herberg, Werner, 88, 92, 97
hilltop complexes, 150–2, 151–3, 155–7, Jalawar, 292
156–68, 192 Jam-Firuzkuh, 2–3, 3, 18, 62, 67–8, 80–1,
Hind, 237 84–7, 85, 111, 115–16, 132–47, 138,
Hindu Kush, 83, 155, 157 140–5, 163, 167, 173–5, 190–7, 217–22,
Hindu-Shahis, 285, 304 218, 254, 267, 321, 322, 338–9; see also
Hinduism, 56–7, 279, 282, 285–9, 304 Firuzkuh Shansabanis
historical inscriptions, 51, 144–5, 204, 304, Jam-Firuzkuh minar, 3, 18, 67–8, 85, 87,
307 111, 133, 137, 143–6, 144–5, 190–3, 197,
historiography 217–20, 218, 221, 254, 321, 322, 339
art-architectural, 17–18, 81–2, 148, 241 Jam-Firuzkuh mosque, 87, 135, 137, 138,
historiographic expectations, 44, 61, 142–3, 146, 221, 339
62 Jam Jaskars Goth, 294, 294–8
Islamic, 15–16 Jam Rud, 85, 115, 132, 137
and nomadic groups, 20–1, 40–3, 174, Jauhar Malik, 194–5, 201, 210
205 Jayaprrcha, 298
oral history, 40, 96, 205 Judaism, 43, 86, 107, 113, 115–16, 139, 142
see also textual sources Jurwas, 41
Hodgson, Marshall, 44 Juzjan Province, 23–7
House of Lusterware, Ghazna, 242, 247 Juzjani, Minhaj al-Din Siraj, 20, 42–4,
Hsuan Tsang, 148 61–7, 86–7, 90, 93, 100, 107–8, 112,
Hudud al-Alam, 41 115–16, 132, 134–40, 143, 146, 154–5,
Hurr Malika, 194 171, 193–4, 197, 237–8, 242, 279, 283,
Husam al-Din ‘Ali Abu al-Hasan, 154 289
Hussain, Talib, 277
Kabul, 82, 83, 148, 149, 169, 239
Ibn al-Athir, 42, 43, 63 Kachh coast, 295
Ibn Karram, 198 Kakrak, 148
Ibrahim ibn Ardashir, 194 Kalar temple, Panjab, 288
iconography see decoration; epigraphy Kalat, 292, 323, 324
‘idgahs, 144, 145, 146 al-Kamil fi al-Tarikh (Ibn al-Athir), 42,
ilkhandids, 66 43, 63
Iltutmish, 282, 311–12 Karramiyya, 114, 146, 194, 197–9, 205–6,
Indus region, 5, 14, 22, 27, 40, 46, 67, 214, 217, 219–20, 339
79, 190, 239, 262, 264, 277–328, 338, Kashi, 63, 134, 154
340–1 Kashmir, 287, 304
inheritance, 61, 174–5 Kevran, Monik, 278
inscriptions see epigraphy Khamseh confederacy, 240
interlocking Kufic script, 202, 209 Khan, Ahmad Nabi, 278, 313
Isfahan, 48–9, 49–52, 51, 116, 237, 260 khanaqahs, 199, 215–17, 221–2, 300
Islam, 8, 15–17, 43–8, 79–80, 108, 112–16, Kharan, 292, 323
145–6, 171–2, 191–3, 197–9, 205–7, Khazars, 46, 113
210, 214–22, 282–3, 290, 293, Khirgai market, 84–6
295–306, 309, 317, 338–9 Khusrau Shah, 64, 67, 263
Islamic historiography, 15–16 Khwabin, 41
Islamization, 17, 43, 113–16, 192, 290, 339 Khwaja Siyah Push, 109, 112
Isma‘ili Shi‘ism, 217, 282–3, 300 Khwarazm-shahs, 26, 42, 45, 46, 65, 154
Istiya, 134, 238 Khwash Rud, 87

Index 397
Khyber-Pakhtunkhwa, 284, 285, 286, 316 endowments, 282, 283

Killigan, 160 patronage, 2, 155, 194–5, 199, 210, 220
King, David, 200 Shah-i Mashhad madrasa, 2, 190–210,
kingship, 16–17, 21–2, 42, 44–60, 106, 196, 200, 202–6, 220, 221, 253, 260,
115, 174, 191, 198, 221, 267–8, 338, 339, 307–8, 313, 339
341 Mahmud Ghaznavi, 16, 41, 53, 84, 114, 154,
Kirminiyya, 310 198, 249, 278, 279–80, 282–3, 285
knowledge, 8–10, 83, 205, 221–2, 264, 298, Mahmud, M. S. 41–2
303 Maimana, 82
Kohzad, Ahmad Ali, 83, 84, 96 Makran coast, 284, 289–98, 302, 303, 323–4
Kufi, ‘Ali, 79 Male Alau see Muna Ala
Kufic script, 25–6, 49, 102, 111, 194, 202–3, Malik Shah, 48
203, 205, 209–11, 214, 215, 219, 244, Mandesh, 84, 86
257, 304, 318, 319 marble, 6, 242, 244, 247, 263
Kuh-i Khara, Jam-Firuzkuh, 139, 142 Mari Indus temple, Panjab, 289
Kukhars, 67 markets, 84–6, 221; see also commerce;
Kurraman, 65, 239 trade
Kushanas, 4–5, 341 marriage alliances, 107, 175, 194
Kushkak, 115 Maru–Gurjara style, 296–8, 302, 303, 306,
Kushk-i Sultan, Jam-Firuzkuh, 135–6, 316–18, 327
139–40, 146 Marv, 51, 65, 116
al-Marvarruzi, Fakhr al-Din
labor, 13, 20, 22, 23, 113, 115, 136, 143–4, Mubarak-shah, 43
167, 191, 195, 222, 264, 281, 302, 306, masjid-i jami’ see Congregational Mosques
339 Mas‘ud I, 41, 48, 249, 279, 280, 282
Lahore, 26, 44, 65, 67, 79, 217, 237, 282, Mas‘ud III, 109, 164, 208–9, 243
283, 237 Mas‘ud III minar, Ghazna, 58, 109–11,
Lahore Fort, 282 110, 208–9, 208–10, 243, 260; see also
Lahore minar, 282 minarets of Ghazni
Lahori Bandar, 290 Mas‘ud III palace, Ghazna, 6, 58, 164,
Lal Mara Sharif, 278, 302, 313, 320, 320–4, 242–7, 244–7, 257–9, 262–3, 265,
322, 325–6, 327–8 279, 323
language, 41, 47 material culture, 20–1, 44, 61, 108, 139, 155,
Lasbela, 292, 293 158, 162–3, 172, 242
Lashkari-Bazar see Bust-Lashkari Bazar Maulana Siraj-i Minhaj, 136, 155
Le Berre, Marc, 150, 157 Mecca, 171, 201
legibility, 145, 204–5, 219 medallions, 25, 212–13
legitimacy, 18, 46, 106 Meister, Michael, 22, 285–9
literacy, 205–6, 219 Mele Hairam, 100
literary production, 42, 53, 136, 154, 192, mercantilism see commerce; trade
221, 282 metalwork, 21
lusterware, 242 migrations, 20, 47, 60, 86–7, 114, 116,
222, 269, 292, 323–4, 339; see also
madhhabs, 114, 193, 197–201, 205–6, 207, nomadism
210, 214–17, 219–20, 278–9, 339 mihrabs
Madin, 134, 154 Barfak II, 157, 164–5
madrasa complexes Chisht domed structures, 190, 212–13,
Afshin, 194 214, 216
Chisht (possible madrasa complex), Jam-Firuzkuh minar, mihrab motif,
190–7, 210–17, 211–13, 215–16, 220, 144–6, 145
221 Lal Mara Sharif, 320

398 Iran to India
mihrabs (cont.) Chhoti masjid, Bhadreshvar, 295, 318

Mosque of the Forecourt, Lashkari Congregational Mosque, Herat, 4, 11,
Bazar, 255, 263 12, 211
Oratory F, Lashkari Bazar, 255, 260, 261 Congregational Mosque, Isfahan, 48,
qibla orientation, 145–6, 199–201, 212, 49–50
214, 278–9 Congregational Mosque, Jam Jaskars
Rajagira mosque, Udegram, 287 Goth, 294, 294–5
riba† of ‘Ali ibn Karmakh, Multan, Congregational Mosque, Multan, 282,
304–6, 306–8, 309, 310–11 283
Sarkhushak, 160 constructed on temple sites, 282
Sin mosque, 53 Dibul-Banbhore, 292
tomb of Muhammad ibn Harun, endowments, 282, 283
Lasbela, 293 Jam-Firuzkuh mosque, 87, 135, 137, 138,
mina’i ware, 139 142–3, 146, 221, 339
Minaret of Jam Archaeological Project Mosque of Kaman, Rajasthan, 10
(MJAP), 18, 136–9, 142–3 Mosque of the Forecourt, Lashkari
minarets of Ghazni, 28, 58, 109–11, 110, Bazar, 255, 258, 263
208–9, 208–10, 243, 260 north Indian duab, 4, 9, 10
minars Qutb mosque, Delhi, 311
Chihil Dukhtaran, Isfahan, 48–9, 50–2 Rajagira mosque, Udegram, 285, 286, 287
commemorative function, 67, 143, 146, Sarkhushak, 160–1
209–10, 218–19, 254, 282, 322 Shahr-i Ghulghula, 163
Da Qadi minar, 99–100, 100, 108–12, Sin mosque, 49, 53
113, 143, 191, 281, 339 Thamban Wari masjid, Jam Jaskars
Daulatabad minar, 52, 99, 102, 110–11 Goth, 294, 294–6, 298
Ghaznavid architecture, 54–9, 98, 109, Umayyad Mosque, Damascus, 48
218 Mughals, 15
Gar minar, 49 Muhammad ibn Qasim, 282
Jam-Firuzkuh minar, 3, 18, 67–8, 85, 87, Muhammad ibn Suri, 84
111, 133, 137, 143–6, 144–5, 190–3, 197, Mu‘in al-Din Sijzi Chishti, 217
217–20, 218, 221, 254, 321, 322, 339 Mu‘izz al-Din, 62–7, 79, 132, 135–7, 173,
Lahore minar, 282 175, 219, 237–49, 254, 259–67, 278–85,
Mas‘ud III minar, Ghazna, 58, 109, 289–92, 298–9, 313, 316, 325–7, 340
208–9, 208–10, 243, 260 al-Mu‘izz li-Din Allah, 282
minarets of Ghazni, 28, 58, 109–11, 110, Mu‘jam al-buldan (Yaqut), 42
208–9, 208–10, 243, 260 Mularaja II, 67
Qasimabad minar, 53, 99, 101, 110–11 Multan, 64, 67, 79, 261, 277–8, 282–3, 289,
Qutb minar, Delhi, 109 300–13, 301, 305–16, 316, 321, 327, 328
Saghar minar, 80 Muna Ala, 91, 98, 99, 102, 103, 104
Saljuq architecture, 50, 98, 109, 218 mural paintings, 152, 156, 161, 267, 268
Sin, 49, 53–4 Murghab River, 197, 200
stellate construction, 56–9, 109, 110
mobility, 18, 23, 45, 51, 60, 112–13, 279, 338 Nab, battle of, 63, 65
Mongols, 10, 44, 82, 116, 137, 139, 155, 241, Najimi, A. W., 195, 201, 203–4
242, 278, 322 Nandana temple, Panjab, 291
Mosque of the Forecourt, Lashkari Bazar, Nasir al-Din Muhammad, 134, 154
255, 258, 263 Nasir al-Din Qabacha, 79
mosques naskhi script, 102, 111, 201, 202, 204, 206,
Arhai Din-ka Jhonpra mosque, Ajmer, 9 211, 244, 304, 313–14
Bara Gumbad mosque, Delhi, 310 Nigar, 148
Barsiyan mosque, 260 Nimruz, 53

Index 399
Nishapur, 51, 52, 65, 114, 194, 197, 198, North Palace, Lashkari Bazar, 270
214–15, 220 patronage, 136, 155, 174, 191
Nizam al-Mulk, 48 Sarkhushak, 6, 155, 160–1, 166–8, 167,
Nizamabad, 101, 104 192
Nizami ‘Aruzi (al-Samarqandi), 42 Shahr-i Ghulghula, 163–4, 168, 339
nomad–urban continuum, 14, 18, 20, 27, Shahr-i Zuhak, 5, 150, 151–3, 155, 162,
31, 47, 53, 86, 94–6, 106–8, 113, 167–8, 192, 339
174, 195, 205, 221, 240, 269–70, 281, South Palace, Lashkari Bazar, 164, 242,
338, 341 254–62, 255–61, 263, 265–7, 266–9,
nomadic studies, 21, 108, 116, 240 308–9
nomadism, 18–21, 40–3, 60–1, 80, 83–7, symbolic function, 136, 140
90, 94, 106–8, 113–17, 133, 136–47, textile palatial architecture, 139–42, 146
167–75, 191–3, 205–6, 221–2, 237, 240, Panjab, 5, 64, 67, 79, 278, 282, 284, 285–7,
269–70, 280, 281, 292–3, 301, 321–4, 288–91, 313, 316, 320, 320–4, 322,
338–9, 341 325–6, 327
north Indian duab, 4, 5–8, 7–10, 14, 40, Panjgur, 292
59–60, 65–7, 79, 190, 218, 237, 250, Parwan/Farwan, 62
262, 277, 280, 282, 303, 325, 338, patronage
340–1 of artisans and labor, 21, 23, 27, 108,
North Palace, Lashkari Bazar, 270 136, 143, 191, 221, 281, 298, 302, 306,
numismatics see coinage 328, 339
cultural, literary and intellectual
Ögedey, 137 patronage, 22, 45, 47, 53, 154–5, 198,
Oghüz, 51, 63, 64, 65, 132, 133, 167, 173, 195, 221
210, 219, 238–9, 241, 254, 263, 322 elite patronage, 22–4, 44, 45, 47–54, 106,
Old Qandahar see Tiginabad/Old 111–13, 136, 154–5, 174, 191–3, 219–22,
Qandahar 327
Old Qandahar cemetery, 171–2, 172 and epigraphy, 192, 204–8, 210, 219–20,
O’Neal, Michael P., 63 262, 264, 307–13
oral history, 40, 96, 205 and kingship, 44, 45, 47–54, 106, 112–13,
oratories 154–5, 174, 198, 338
Robat Sharaf, 54, 56 of madrasa complexes, 2, 155, 194–5, 199,
South Palace, Lashkari Bazar, 255, 207, 210, 220
259–61, 259–62, 263, 265, 309 non-elite patronage, 23, 112, 299
Orientalism, 29 of palaces, 4, 136, 155, 167, 174, 191
origin myths, 42–3, 107–8 Saljuq patronage, 48–53, 109, 144, 199
ornament see decoration and sectarian affiliation, 193, 197–9, 210,
overflowing pot motifs (purnaghata), 295, 219–20, 339
305, 308, 310, 315–18, 318 of temples, 285
of tombs, 299, 301, 302, 306–13, 321–2,
paganism, 113, 114 327
‘Palace of Racquets’, Lashkari Bazar, 255, paved floors, 6, 138, 142–3
264–5, 265 Peshawar, 67, 283, 284–5
palaces photographic documentation, 29–31, 88,
Ghaznavid architecture, 163–5, 193 91, 109, 195
Kushk-i Sultan, Jam-Firuzkuh, 135–6, Pigeon House, Lashkari Bazar, 270
139–40, 146 pilasters, 305, 310, 315, 318
Mas‘ud III palace, Ghazna, 6, 58, 164, pilgrimage, 51, 62, 169, 239, 282, 285,
242–7, 244–7, 257–9, 262–3, 265, 299–300
279, 323 Pinder-Wilson, Ralph, 208–9
north Indian duab, 4, 7, 8 pishtaqs see portals

400 Iran to India
portals, 50, 143, 146, 203, 203, 204, 310 rahmana-prasada, 298, 306
Powell, Josephine, 29, 88 Rajagira mosque, Udegram, 285, 286,
prestige cultures, 8, 44–60, 112–13, 338 287
proselytism, 113, 114, 148, 198–9, 205–6, Rajasthan, 9–10, 67, 296–8, 302, 316
214, 300, 339 Rayy, 51
al-Razi, Fakhr al-Din ibn ‘Umar, 154
Qaitul Buddhist complex, 169–71, 170 Rehman, Abdul, 277
Qala‘-yi Chahar Baradar, 98, 98 re-inscription, 263–5, 280–4, 299, 321,
Qala‘-yi Gauhargin, 158 325–8, 340
Qala‘-yi Zarmurgh, 99, 108, 113, 116, 143, renovations, 3–4, 48, 51, 104, 210, 211, 237,
281 242–7, 255–69, 303, 305, 308–9, 321,
Qandahar see Tiginabad/Old Qandahar 325
Qarakhanids, 26, 45–6 reoccupation, 155, 162, 165, 168, 174, 242,
Qarakhitay, 26, 45–6, 167 249, 263–4, 325
Qashqai confederacy, 240 repurposing, 81, 96, 104, 106, 108, 133,
Qasimabad minar, 53, 99, 101, 110–11 155–60, 162, 165, 174, 192, 193, 263,
Qasr-i Zarafshan, Jam-Firuzkuh, 139, 281
140–1 residential architecture, 137–9, 248
qibla orientation, 145–6, 199–201, 212, 214, revetments, 157–8, 165, 247, 255, 260,
278–9 262–3, 267
Qunduz, 148, 150 riba† of ‘Ali ibn Karmakh, Multan, 277,
Qur’an 301, 302–3, 304–6, 304–11, 309,
commentary (tafsir), 197 327
illuminated texts, 197, 199 Risala-yi Baha’iyya (al-Razi), 154
Qur’anic inscriptions, 26, 144–5, 193, ritual, 106, 114, 150, 199, 287, 289, 296,
197, 203–4, 207–14, 212–13, 215, 303
218–19, 222, 251–4, 252–3, 259–62, Robat Sharaf, 51–2, 54, 55–7
260, 302–13, 313–14, 327 robber holes, 136, 139
Surat al-Baqara (Q. II), 214, 251, 254, Roman empire, 10
260 royal enclaves, 248–54
Surat al-Fath (Q. XLVIII), 203, 204, royal genealogies, xviii, 42–3
208–10, 259–61, 307–9, 313 Rud-i Ghur, 87
Surat al-Fatiha (Q. I), 309 al-Rukkhaj, 46, 105, 169, 171, 173–4, 300
Surat al-Hashr (Q. LIX), 203, 207, 210, Rutbils, 105–6, 149, 169, 239
214
Surat al-Ikhlas (Q. CXII), 214, 260, 261, Sabuktigin, 16, 53, 114, 153–4, 221, 248
309, 312 Sadr al-Din ‘Ali Haisam, Imam, 194
Surat al-Imran (Q. III), 214 Safavids, 48
Surat al-Kauthar (Q.CVIII), 310, 312 Saffarids, 105, 153
Surat Maryam (Q. XIX), 144–5, 146, Saghar minar, 80
197, 219 Sahih (al-Bukhari), 207
Surat al-Rahman (Q. LV), 309–10 Saif al-Din, 61, 62, 64, 134, 195, 201, 207,
Surat al-Saf (Q. LXI), 218 238
Surat al-Shu’ura (Q. XLII), 310–12 Saljuq ibn Duqaq, 47
Surat al-Tauba (Q. IX), 309 Saljuqs, 8, 19, 40, 42, 44–53, 59–64, 98,
Throne verse, 214 106, 109, 114–16, 139–42, 144, 154, 167,
Qutb al-Din Aibeg, 65, 311 195, 197, 199, 218, 237, 290–2, 322–3,
Qutb al-Din Muhammad, 62, 132, 134–5, 338
143, 194 Samanids, 16, 53–4, 153–4, 221, 248
Qutb minar, Delhi, 109 al-Samarqandi, Nizami ‘Aruzi, 154
Qutb mosque, Delhi, 311 Sana’i, 1

Index 401
Sanga, 134 Sourdel-Thomine, Janine, 22, 26, 144–5,

Sanjar, 51, 63, 65 217–18, 219, 253, 254–9, 262
saraparda, 139–42 South Palace, Lashkari Bazar, 164, 242,
sard-sir encampments, 18, 68, 84, 87, 105, 254–62, 255–61, 263, 265–7, 266–9,
116, 133, 136–7, 139, 143, 146, 190, 194, 308–9
217, 222, 237, 321, 339 squinches, 161, 287, 304, 315
Sar-i Pul, 82, 86 stepwells, 250
Sarkhushak, 6, 155, 160–1, 166–8, 167, stucco, 26, 50, 100–3, 102–4, 111, 157, 212,
192 216, 242–3, 245–6, 251, 255, 259–60,
Sasanians, 31, 48, 53, 81, 100–6, 109, 149, 260, 267
150, 153 stupas, 58–9, 59, 169, 170
Sceratto, Umberto, 322 Sufism, 214–17, 220–1, 299–300, 339
Schlumberger, Daniel, 255, 265 Sukkhur, 278, 302, 313–18, 317–18, 323,
Second Anglo-Afghan War, 29 327, 328
sectarian affiliation, 193, 197–201, 204, 207, summer capitals see sard-sir encampments
210, 214–17, 219–20, 278–9, 339 Sunni Islam, 198, 207, 210, 219–20, 283
sedentism, 18, 20, 60, 94, 108, 113, 146–7, al-Surabadi, Abu Bakr ‘Atiq ibn
167–8, 191, 237, 267–70, 339, 341 Muhammad, 197
Shafi‘i madhhab, 197–8, 199, 201, 214, 217, Swat, 283, 284–5
278, 339
Shahada, 198, 206, 213 Tabaqat-i nasiri (Juzjani) see Juzjani,
Shah-i Mashhad madrasa, 2, 190–210, Minhaj al-Din Siraj
196, 200, 202–6, 220, 221, 253, 260, Taimanis, 94
307–8, 313, 339 Taiwara, 87, 93, 95
Shah-nama, 42 Taj al-Din Yildiz, 65
Shahr-i Ghulghula, 29, 150–1, 155, 156–7, Taj al-Mulk, 48
161–5, 192, 339 takbir of tashriq, 203, 204, 207
Shahr-i Zuhak, 5, 150, 151–3, 155, 162, 192, Takht-i Bahi Buddhist complex, 284
339 Tapper, Richard, 21
Shaivite Hinduism, 285, 289 Taq-i Bustan, 103, 105
Shams al-Din see Ghiyath al-Din Tara‘in, battle of, 311
Shams al-Din Nasri, 238, 267 Tarikh-i Sistan, 106
Sharada script, 304 Tarzi, Z., 150
Sharaf al-Din Abu Tahir ibn Sa‘d al-Din temples
ibn ‘Ali al-Qummi, 51 Aditya temple, Multan, 282, 300
Shihab al-Din see Mu‘izz al-Din Ambh Sharif temple, Panjab, 290–1
Shi‘ism, 198, 204, 207, 217, 282–3, 300 Bhodesar temple, Nagarparkar, 297
Shish ibn Bahram, 107 Gandhara–Nagara school, 285–9,
Shishanis, 90, 107–8, 195 288–91, 302, 303, 305–6, 309, 327
Shrine of Ibrahim, Bhadreshvar, 296, 319 Kalar temple, Panjab, 288
Sialkot, 67 Kashmiri, 287
Sin minar, 49, 53–4 Mari Indus temple, Panjab, 289
Sin mosque, 49, 53 Nandana temple, Panjab, 291
Sindh, 67, 278, 282–4, 289–98, 292, 294, patronage, 285
297, 302, 303, 313–18, 317–18 plunder and destruction, 279–80, 282
Sistan, 52, 53, 56, 61, 65, 83, 87, 99, 101, tents, 94–5, 95, 139, 142, 222, 339, 340
104–6, 109–11, 111–12, 114, 153, 238, Tepe Hissar, 100, 101, 103
267 Tepe Sardar, 56–9, 59
skilled labor see artisans terracotta, 25, 242–7, 292–3, 293, 324
slavery, 42, 53 textile enclosures, 139–42, 146
Somnath campaign, 280 textiles, 21, 84–6, 139–42, 247

402 Iran to India
textual sources, 2, 17, 21, 40–4, 61, 80, 82, trading routes, 19, 41, 51, 150, 153, 169–71,
84, 107–8, 134–7, 146–7, 174, 193–4, 173, 175, 239, 282, 289–90
199, 239, 242 transhumance see nomadism
Thamban Wari masjid, Jam Jaskars Goth, Transoxiana, 46, 310
294, 294–6, 298 trefoil arches, 287, 288–90, 303–5, 306–8,
Thomas, David, 18, 109, 110, 136–7 313, 316
thuluth script, 202, 203, 204 tribute, 41, 61, 63
Tiginabad/Old Qandahar, 3, 4, 28, 82, Tughril, 48
132–3, 169–75, 170, 172, 191–2, 238–9, Tukharistan, 149, 154, 156, 167
242, 248, 300–1
Timurids, 8 Uchh, 67, 79
tombs Udabhandapura, 285
of Ahmad Kabir, Multan, 277–8, 301, Udegram, 285, 286, 287
302–4, 309–10, 314–16 Umayyad Mosque, Damascus, 48
of ‘Ala’ al-Din Muhammad, Bistam, Umayyads, 15, 104, 114, 169, 290
65, 66 unbaked brick, 90–1, 93
domed cube structures, 23–4, 24, 213, UNESCO world heritage sites, 147–8
278, 293, 302, 304, 314, 315, 317, 320, Uruzgan, 82
320, 324, 326 al-‘Utbi, 41
of Ghiyath al-Din Muhammad ibn
Sam, Herat, 4, 13, 14 vaulting, 12, 66, 152, 168
of Iltutmish, Delhi, 311–12 vegetal ornament, 144–5, 157, 162, 295, 305
Indus region, 277–8, 299–324, 301, vernacular architecture, 23, 112, 328
305–20, 322, 325–6 victory commemoration, 67, 143, 146,
Lal Mara Sharif tombs, 278, 302, 313, 209–10, 218–19, 253–4, 282, 322
320, 320–4, 322, 325–6, 327–8 viharas, 169
of Mir Sayyid Bahram, Kirminiyya,
310 Wajiristan, 134
of Muhammad ibn Harun, Lasbela, Warshada, 134
292–3, 293, 324 watch towers, 93
patronage, 299, 301, 302, 306–13, 321–2, wells, 249–51, 250–1
327 Western Turks, 31, 47, 81, 106, 149–51, 153,
riba† of ‘Ali ibn Karmakh, Multan, 277, 156, 157, 167
301, 302–3, 304–6, 304–11, 309, 327 winter capitals see garm-sir encampments
of Sadan Shahid, Multan, 261, 277–8,
301, 302–4, 309, 311–13 Yakaulang, 158, 160–1
of Salar Khalil, Juzjan, 23–7, 24, 25 Yaman, 89, 96, 96
Sarkhushak, 160–1, 167 Yamini-Ghaznavids see Ghaznavids
Shah-i Mashhad madrasa possible tomb, Ya‘qut, 42
190, 195, 201, 202, 207 yurts, 94–6, 95
of Shahzada Shaikh Husain, Bust, 23–5,
26, 27 Zabulistan, 16, 43, 46, 105, 109, 149, 169,
of Sheikh Yusuf Gardizi, Multan, 301 221, 239–41, 243, 248, 264, 285, 325,
Shrine of Ibrahim, Bhadreshvar, 296, 327, 340
319 Zamindawar, 16, 46, 63–4, 82–3, 87, 105–6,
Sukkhur tombs, 278, 302, 313–18, 114, 139, 154, 169–74, 194, 221, 239–43,
317–18, 323, 327, 328 247–50, 255, 264, 282, 300, 325, 327,
see also burial mounds; graves 340
trade, 20, 21, 41–2, 84–6, 113–16, 142, 153, Zaranj–Nad-i Ali, 53, 61, 109, 111, 238
169–73, 220, 239–40, 282–3, 289–90; Ziad, Waleed, 279
see also commerce zoomorphic motifs, 247, 304, 323


Alka Patel - Iran To India - The Shansabānīs of Afghanistan, C. 1145-1190 CE-Edinburgh University Press (2021)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Alka Patel - Iran To India - The Shansabānīs of Afghanistan, C. 1145-1190 CE-Edinburgh University Press (2021)

Uploaded by

Copyright:

Available Formats

IRAN TO

PATEL 9781474482226 PRINT.indd 1 14/09/21 5:36 PM

PATEL 9781474482226 PRINT.indd 3 14/09/21 5:36 PM

© Alka Patel, 2022

Edinburgh University Press Ltd

Typeset in 11/13 Adobe Garamond by

ISBN 978 1 4744 8222 6 (hardback)

Published with the support of the University of Edinburgh Scholarly Publishing

Introduction: The Elephant and its Parts 1

Appendix: Shansabānī Religious and Historical Inscriptions in

PATEL 9781474482226 PRINT.indd 5 14/09/21 5:36 PM

I.1 Remains of Shah-i Mashhad madrasa, Gharjistan 2

PATEL 9781474482226 PRINT.indd 6 14/09/21 5:36 PM

1.1 Congregational Mosque, Isfahan, Iran, tenth century ce;

PATEL 9781474482226 PRINT.indd 7 14/09/21 5:36 PM

2.15 Qala‘-yi Chahar Baradar, Ghur Province 98

PATEL 9781474482226 PRINT.indd 8 14/09/21 5:36 PM

3.21 Barfak II/“Danestama:” mihrab from southeast 164

PATEL 9781474482226 PRINT.indd 9 14/09/21 5:36 PM

5.6 Ghazni, palace of Mas‘ud III (1099–1115): dado remnants 246

PATEL 9781474482226 PRINT.indd 10 14/09/21 5:36 PM

6.12 Chhoti masjid, Bhadreshvar, Kachch District, Gujarat, India;

PATEL 9781474482226 PRINT.indd 11 14/09/21 5:36 PM

PATEL 9781474482226 PRINT.indd 12 14/09/21 5:36 PM

finalization and submission to a press for review. It is worth stating in no

PATEL 9781474482226 PRINT.indd 13 14/09/21 5:36 PM

PATEL 9781474482226 PRINT.indd 14 14/09/21 5:36 PM

PATEL 9781474482226 PRINT.indd 15 14/09/21 5:36 PM

PATEL 9781474482226 PRINT.indd 16 14/09/21 5:36 PM

PATEL 9781474482226 PRINT.indd 1 14/09/21 5:36 PM

These volumes are about encountering a proverbial elephant – here

Figure I.1 Remains of Shah-i Mashhad madrasa, Gharjistan (Badghis Province,

PATEL 9781474482226 PRINT.indd 2 14/09/21 5:36 PM

Jam-Firuzkuh in central Afghanistan (Figure I.2) long remained enigmatic

PATEL 9781474482226 PRINT.indd 3 14/09/21 5:36 PM

PATEL 9781474482226 PRINT.indd 4 14/09/21 5:36 PM

PATEL 9781474482226 PRINT.indd 5 14/09/21 5:36 PM

PATEL 9781474482226 PRINT.indd 6 14/09/21 5:36 PM

PATEL 9781474482226 PRINT.indd 7 14/09/21 5:36 PM

are particularly noteworthy. First, during the 1190s ce the Shansabanis’

PATEL 9781474482226 PRINT.indd 8 14/09/21 5:36 PM

technologies and modes of communication constantly becoming available.

PATEL 9781474482226 PRINT.indd 10 14/09/21 5:36 PM

Figure I.13 Congregational Mosque, Herat, interior: western iwan, vault.

monolithic, or even unidirectional. Not only has empire-building always

PATEL 9781474482226 PRINT.indd 12 14/09/21 5:36 PM

PATEL 9781474482226 PRINT.indd 13 14/09/21 5:36 PM

PATEL 9781474482226 PRINT.indd 14 14/09/21 5:37 PM

Such a close examination of approximately eight–nine decades in the

PATEL 9781474482226 PRINT.indd 15 14/09/21 5:37 PM

only grown in appeal and practitioners. And, as if echoing the aforemen-

PATEL 9781474482226 PRINT.indd 16 14/09/21 5:37 PM

Chapters 5 and 6). By contrast, rather than assuming largely unprob-

Although ultimately ephemeral (and thus largely ignored by historians and

PATEL 9781474482226 PRINT.indd 17 14/09/21 5:37 PM

While this description of the Shansabanis is – again – not strictly

[n]omads can be designated more concretely as pastoralists – people at some

Following Crossley, then, for convenience I use the less cumbersome

PATEL 9781474482226 PRINT.indd 18 14/09/21 5:37 PM

these undoubtedly took place in response to a variety of social, economic,

In conditions of wide access to resources and simple technology, small

I have found such an apprehension of the Shansabanis’ transformation

If rich barbarians successfully resist both their state-organized and unorganized